Abstract
Malnutrition is a disease that imposes a significant healthcare cost burden in the United States, especially when left undiagnosed and untreated for an extended period of time. This article discusses traditional malnutrition diagnostic criteria for adults and why registered dietitian nutritionists and physicians should no longer use these criteria to determine nutrition status. It concludes with the malnutrition clinical characteristics currently accepted in the United States and globally, with implications for practice. Clinical documentation specialists and medical coders can use this information to better interpret medical record documentation and assign the correct International Classification of Diseases, 10th Revision, Clinical Modification codes to the coding abstract.
Keywords
Introduction
Malnutrition (undernutrition) results from inadequate intake or assimilation of nutrients that are required to promote growth and prevent chronic or acute disease, and is often characterised by stunting, wasting, underweight and micronutrient deficiencies (World Health Organization (WHO), 2016). Poor nutrient intake and diseases or conditions that cause hypermetabolism or malabsorption are the main factors leading to malnutrition. Compared to well-nourished patients, children, adolescents and adults with malnutrition are more likely to be admitted to a hospital or other healthcare facility and they incur higher costs, resulting from an extended length of stay and greater nursing and ancillary staff time required to provide care (Fingar et al., 2016). Malnutrition is more prevalent in adults, with almost one-third of the costs associated with this condition occurring in those aged 65 and older (Fingar et al., 2016).
Although the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) system includes malnutrition codes (see Box 1), inconsistent use of criteria to identify the presence and degree of malnutrition creates challenges in assigning an appropriate code based on medical record documentation (White et al., 2012). This conundrum makes conducting and accurately interpreting health economics, epidemiology and outcomes research related to malnutrition extremely difficult. It is impossible to accurately determine the prevalence, costs, complications and effective treatments for malnutrition if diagnostic malnutrition criteria are applied inappropriately or inconsistently in daily provision of care in healthcare facilities and ambulatory care settings, or in medical coding for claims forms and research.
ICD-10-CM malnutrition codes.
MCC: major complication or co-morbidity; ICD-10-CM: International Classification of Diseases, 10th Revision, Clinical Modification; CC: complication or comorbidity.
Therefore, healthcare professionals need to use a consistent approach to diagnose malnutrition, and with the assistance of documentation specialists, document the malnutrition diagnosis in the medical record so that the correct ICD-10-CM code is assigned to claims forms. This article discusses traditional malnutrition diagnostic criteria for adults and rationale for no longer using these criteria to determine nutrition status. It concludes with the malnutrition clinical characteristics that have been in use in the United States since 2012 and the global definitions of malnutrition, with implications for practice.
Non-evidence-based malnutrition criteria
Body mass index
Body mass index (BMI) is a ratio of weight to height, developed to estimate (not definitively determine) body fat and risk for developing chronic disease; a higher BMI indicates a higher risk for obesity-associated diseases like heart disease, some cancers, diabetes and others (National Heart, Lung, and Blood Institute, 2018). Its use as an indicator for health risk and population health intervention is controversial, as it can overestimate body fat composition in persons with a high muscle mass (such as athletes) and underestimate health risk in those with low muscle mass (such as persons with malnutrition or sarcopenia, which is a loss of muscle mass associated with ageing) (Centers for Disease Control and Prevention, 2018). A BMI measurement also does not distinguish the distribution of body fat in the visceral or subcutaneous body compartments, which has significant implications on chronic disease morbidity and mortality (Campbell et al., 2018). Furthermore, there may be nutritional deficiencies and undernourishment in those who are overweight or obese due to the overconsumption of processed, calorie-dense foods with low nutrient value (Astrup and Bugel, 2019).
BMI is a diagnostic criterion often used in countries other than the United States to identify and classify malnutrition, and has been recommended for use in the global diagnostic criteria published in 2018 (discussed further in a subsequent section) (Jensen et al., 2019). Although some healthcare providers still use BMI as a diagnostic criterion for malnutrition in the United States, Jensen et al. (2019) list the caveat that BMI is not intended for use in the United States since North Americans are more likely to be overweight (BMI greater than 25) or obese (BMI greater than 30). A person would not meet the BMI cut-offs listed in that global diagnostic criteria until a substantial amount of weight was lost, thereby potentially missing a malnutrition diagnosis until it was too late to effectively treat it. Similarly, some people have a smaller bone structure and frame size with a genetic predisposition to be thin (Frisancho, 1984). These people may have a BMI less than 18.5 (the cut-off value considered to be underweight by the National Institutes of Health) but are not malnourished. Therefore, BMI should be included in nutrition assessment documentation as one part of the overall evaluation of health status and risk of chronic disease, including malnutrition. It should not be used as the sole criterion to diagnose malnutrition, and clinicians need to be aware that malnutrition can occur at any BMI.
WHO criteria
In 1999, the WHO published the resource “Management of Severe Malnutrition: A Manual for Physicians and Other Senior Health Workers” (WHO, 1999). The manual includes guidelines for identifying and treating malnutrition in children and a brief discussion of adolescents and adults. The WHO recommends assessment of oedema and anthropometric data compared to the National Center for Health Statistics reference population, and severe malnutrition is classified as oedematous, severely wasted or severely stunted. Measurements with a standard deviation score of less than −3 from the median are considered severely malnourished (see Table 1).
Malnutrition classifications as outlined in the WHO’s Management of Severe Malnutrition manual.
WHO: World Health Organization; SD: standard deviation.
In the 60-page manual, only 2 pages are devoted to the discussion of malnutrition in adults and adolescents. Malnutrition is described as “a primary disorder in adolescents and adults in conditions of extreme privation and famine” (WHO, 1999). While other conditions and illnesses are listed as possible contributors to malnutrition, no criteria are specified to distinguish chronic or acute illness or inflammation. The WHO uses BMI as the primary criterion to classify malnutrition in adults and adolescents; the cut-off values used to describe malnutrition are listed in Box 2. In adolescents, the WHO recommends BMI-for-age percentiles with a cut-off value of less than the 5th percentile compared to the reference standard.
Classification of nutrition status based on BMI according to the WHO’s Management of Severe Malnutrition manual.
BMI: body mass index; WHO: World Health Organization.
Since this criterion was created for use in developing countries, it should not be utilised in the United States, where the food supply and healthcare system are substantially different than the countries for which the criteria and associated population health interventions were intended.
Serum protein levels
Historically, serum levels of the hepatic proteins (formerly known as “visceral” proteins), albumin and prealbumin (or transthyretin), have also been used erroneously to determine the presence and degree of malnutrition (JeVenn et al., 2017; Shenkin, 2006; Yeh et al., 2018). Threshold values for indicating malnutrition were even established, with the assumption that improving nutrition intake and resolving malnutrition would increase serum levels of these proteins. However, albumin and prealbumin are negative acute phase reactants, thus decreased levels are indicative of inflammation, not malnutrition (Yeh et al., 2018). Infection, disease and injury cause the liver to increase production of positive acute phase reactants, such as cytokines, and decrease production of albumin and prealbumin. Patient outcomes, such as ventilator dependence, length of hospital stay and mortality are worsened in patients in an inflammatory and catabolic state as indicated by low albumin and prealbumin, which do predict morbidity and mortality (Yeh et al., 2018). However, in the presence of ongoing inflammation or injury, nutrition interventions and adequate nutritional intake will not significantly improve albumin and prealbumin levels, thus these proteins should not be used to diagnose or classify malnutrition or assess the adequacy of nutrition interventions (JeVenn et al., 2017; Shenkin, 2006; Yeh et al., 2018). Additionally, albumin and/or prealbumin may not be decreased in those that are chronically malnourished without inflammation, due to the body’s compensatory response (White et al., 2012).
Evidence-based malnutrition criteria
The need for a more consistent and reliable diagnostic approach for malnutrition led to the formation of an International Consensus Guideline Committee in 2009 to define malnutrition using an aetiology-based approach. The Academy of Nutrition and Dietetics (Academy) and the American Society for Enteral and Parenteral Nutrition (ASPEN) were the nutrition organisations leading this important work to develop a set of criteria for assessing and classifying malnutrition in the United States. This work was finalised for adults in 2012 (White et al., 2012), paediatrics in 2014 (Becker et al., 2015), and neonates in 2018 (Goldberg et al., 2018). The adult malnutrition clinical characteristics delineate criteria to diagnose severe or non-severe (moderate) malnutrition (White et al., 2012). Although ICD-10-CM codes exist for mild, moderate and severe protein-calorie malnutrition, insufficient research exists to distinguish between adult mild and moderate malnutrition (White et al., 2012). Therefore, in this population, the use of the ICD-10-CM code corresponding to mild protein-calorie malnutrition should only be used if the physicians, registered dietitian nutritionists (RDNs) and other healthcare professionals at the facility identify characteristics of mild malnutrition for that population using a consensus-based and research-driven process.
In the diagnostic approach for adult malnutrition, three basic categories are used to describe the aetiology of malnutrition, which is necessary in order to determine the most appropriate treatment. These categories are chronic or acute injury or illness, and social/environmental causes (White et al., 2012). Six malnutrition clinical characteristics are outlined with thresholds for moderate and severe malnutrition, including weight changes, nutrient intake and four physical findings that can be identified by an RDN or physician while conducting a nutrition-focused physical exam. Both the paediatric (Becker et al., 2015) and neonatal (Goldberg et al., 2018) malnutrition classification systems define threshold values for mild, moderate and severe malnutrition without using the aetiology-based approach that is used for adults. As expected, the criteria are focused on appropriate growth using reference standards and nutrient intake sufficient to meet needs for growth and development. This is similar to the approach used throughout the world by the WHO.
The reader is referred to the open access consensus statements for each age group in the Journal of the Academy of Nutrition and Dietetics for further detail. Since 2012, several studies have verified the ability of these malnutrition clinical characteristics to predict patient outcomes (known as predictive validity) (Drake et al., 2018; Hiller et al., 2017; Hudson et al., 2018; Mogensen et al., 2018; Vest et al., 2018). Therefore, clinicians and researchers are encouraged to use the malnutrition clinical characteristics described in these documents to ensure diagnostic consistency between clinicians and facilities. This will enable research to more easily and accurately determine the prevalence of malnutrition, common aetiologies and the most effective interventions to treat this debilitating condition. While the Centers for Medicare and Medicaid Services (CMS) has not formally accepted this classification system, they have not accepted any other classifications or definitions for malnutrition either, and the ICD-10-CM codes listed in Box 1 are still randomly assigned in facilities that have not adopted the consensus statement criteria. Regardless of the malnutrition clinical criteria chosen, each healthcare facility needs to create a policy outlining the diagnosis and classification of malnutrition, and all clinicians should consistently apply these criteria to each patient in order to accurately identify the presence and degree of malnutrition. RDNs, physicians and other healthcare practitioners should still utilise clinical judgement in the diagnosis of malnutrition, while being consistent with the facility-approved parameters. As the malnutrition clinical criteria include physical parameters such as muscle and fat wasting, diagnosing practitioners should be trained to assess these criteria accurately and practice these skills regularly to maintain competence and sound clinical judgement.
The Academy/ASPEN malnutrition clinical characteristics will be formally validated over the next 2 years to ensure widespread adoption as the standardised definition of moderate and severe malnutrition (Hand et al., 2016). A study has been designed, research sites and RDNs were recruited to participate in 2018, and data collection and analysis will occur in 2019. The validation model will measure predictive validity, that is, demonstrating that patients meeting the threshold values of the malnutrition clinical characteristics have poorer health outcomes (Hand et al., 2016).
International malnutrition diagnostic criteria for adults
The Global Leadership Initiative on Malnutrition (GLIM) includes several of the major global clinical nutrition societies, including the United States’ Academy and ASPEN, with a goal to build a global consensus around core diagnostic criteria for malnutrition in adults in clinical settings (Jensen et al., 2019). As mentioned, BMI is one of the malnutrition diagnostic criteria included in the GLIM consensus report, but it is not intended for use in North America.
The Academy/ASPEN criteria share the same goals as the GLIM criteria: Standardise malnutrition diagnoses. Document malnutrition clinical characteristics on patients, clients and community-living people using discrete fields for data extraction to understand the prevalence and scope of malnutrition. Understand common aetiologies for malnutrition and determine the best treatment methods.
The similarities and differences between the two sets of consensus statements for adults are described in Box 3.
Comparison of adult malnutrition clinical characteristics from the GLIM malnutrition diagnostic criteria and the consensus statement criteria from the Academy and ASPEN.
GLIM: Global Leadership Initiative on Malnutrition; ASPEN: American Society for Enteral and Parenteral Nutrition; Academy: Academy of Nutrition and Dietetics; BMI: body mass index.
The GLIM criteria are meant to complement, but not replace, similar medical diagnoses of cachexia, sarcopenia and frailty, which are conditions that are similar to but distinct from malnutrition. The GLIM committee will re-evaluate the criteria every 3–5 years when new research is available, and they are working with the WHO to incorporate the criteria into the ICD-11-CM system when it is released (Jensen et al., 2019).
Healthcare practitioners in the United States are encouraged to use the Academy/ASPEN criteria, as these were developed specifically for the North American population. However, these and the GLIM criteria complement one another. If a healthcare facility chooses to adopt the GLIM criteria as their standardised diagnostic method, they should remove BMI as a criterion.
Conclusion
In the past, health researchers and economists have used differing malnutrition classification systems to study the prevalence of malnutrition and its associated health outcomes. Since large-scale epidemiological studies utilise claims data with ICD-10-CM codes, the lack of a standardised definition makes it impossible to identify which diseases are most likely to lead to malnutrition and determine the most effective treatments. A clearer understanding of the intricacies of malnutrition prevention, diagnosis and treatment would help to allocate healthcare labour and financial resources in the most appropriate manner. Since 2012, RDNs, physicians, non-physician practitioners and other healthcare professionals have used the Academy/ASPEN malnutrition clinical characteristics to more accurately determine the presence and severity of malnutrition. Documentation specialists need to be aware of the appropriate diagnostic criteria for malnutrition when assigning ICD-10-CM codes. Their role is essential in communicating with and educating physicians to ensure that malnutrition is appropriately identified and coded.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
