September 02, 2016
We like to classify just about everything in life for one reason and one reason alone: to help keep us organized. If you need proof, look no further than your beloved smart-phone’s PokémonGo app. PokémonGo would simply not be PokémonGo without the handy-dandy Pokédex to show all your friends what you’ve accomplished by climbing over a few fences and trespassing in a neighbor’s yard or two. The Pokédex has been a crucial part of the Pokémon franchise from its inception. It’s what allows users to categorize hundreds of monsters into their type, gender, weaknesses, abilities, and evolution stage, and is what helps users prepare for battle. Without it, Pokémon would have been hopeless decades ago and Ash Ketchum would have never made it into the big leagues.
A century ago, a few brilliant researchers and physicians convened to logically and systematically organize something a bit more menacing – the causes of illness and death. This ultimately gave birth to what is now known as the International Classification of Diseases (ICD)—medicine’s own Pokédex.
The ICD is a set of standardized codes that categorize causes of morbidity and mortality that can then be used across the world. Throughout its lifespan, it has undergone 10 major revisions (and many more minor ones) to reflect the latest medical knowledge and the detail with which diagnoses are made.
One variant to the ICD – the Clinical Modification-ICD 10 (ICD-10-CM) – is used strictly in the U.S. healthcare setting to classify diseases, injuries, health encounters and inpatient procedures. It is primarily used for billing purposes by insurance companies, although it is sometimes also used for surveillance and research. Currently, it has 71,924 procedural codes and 69,823 diagnostic codes (one very large Pokédex indeed)(1). Though, like any classification system of this magnitude, it is bound to have its weak points.
Let’s examine antibiotic resistance – one of CDDEP’s signature topics. There are several notorious hospital-borne pathogens that are resistant to even the most powerful antibiotics: Enterococcus, Staphylococcus, Klebsiella, Acinetobacter, Pseudomonas, and Enterobacter. Only one of these organisms has a specific code that identifies both the organism and the antibiotic it is resistant to, Methicillin-Resistant Staphylococcus aureus (MRSA). The remaining organisms are identified using a combination of general codes that identify the bug of interest, the class of antibiotic, and resistance. Carbapenem-resistant Enterobacteriacae (CRE), for example, is classified using one code (B96.89) for “other specified bacterial agent as the cause of diseases classified elsewhere,” plus another (Z16.10) for “resistance to unspecified beta lactam antibiotics”(2). The same bug can also be identified using code A48.8 “other specified bacterial diseases,” modified by code Z16.10.
From a research and surveillance perspective, it’s not difficult to deduce why a single specific code is preferable to matching multiple obscure codes. For starters, a single code is easier to work with when performing data analysis. Second, there is a sense of trust in using a specific code with an exact description, as opposed to multiple codes that need to be combined in order to make sense of them. And third, some resistant drug-bug infections (like CRE) have multiple ways to code for the same case, which could easily lead to confusion and miscounting among administrative coders and researchers alike.
In fact, by examining previous general MRSA codes, several studies have already reported that specific codes are more often reliable than general codes (3,4). The authors of these studies concluded that general codes have a tendency of being overused and misused when a patient is classified with an infection. This can happen when administrative coders would code for MRSA infection to patients who had a previous hospital visit, a patient who was being screened for a MRSA infection, or a patient who had prior history of colonization (but not infection) with S. aureus.
Reliable codes are critical for research to accurately understand the patterns of antibiotic-resistant infections and the associated costs, including the need for more expensive antibiotics, multiple courses of antibiotics, increased length of hospital stay, and increased morbidity and mortality (5). So rather than trying to catch all new infections using cryptic codes that have been shown to be inaccurate and worrisome, conversations need to be had that would create new codes and would help us catch ‘em all.
Nestor Mojica is a Research Analyst at CDDEP.
Sources
1. CDC. International Classification of Diseases, (ICD-10-CM/PCS) Transition – Background. Centers for Disease Control and Prevention. 2015.
2. Galuardi G. Superbug Code Alert – Carbapenem – Resistant Enterobacteriaceae (CRE) [Internet]. Libman Education. 2013. Available from: https://www.libmaneducation.com/code-alert-carbapenem-resistant-enterobacteriaceae/
3. Schweizer ML, Eber MR, Laxminarayan R, Furuno JP, Popovich KJ, Hota B, et al. Validity of ICD-9-CM coding for identifying incident methicillin-resistant Staphylococcus aureus (MRSA) infections: is MRSA infection coded as a chronic disease? Infect Control Hosp Epidemiol Off J Soc Hosp Epidemiol Am. 2011 Mar;32(2):148–54.
4. Schaefer M, Ellingson K, Conover C, Genisca A, Currie D, Esposito T, et al. Evaluation of International Classification of diseases, ninth version, Clinical Modification Codes for reporting MRSA infections at a Hospital in Illinois. Infect Control Hosp Epidemiol. 2010 Mar 30;31(5):463–8.
5. Hargreaves J, Kok J. Australian hospital morbidity data on antibiotic resistance. Commun Dis Intell CDI [Internet]. 2003 May [cited 2016 May 23];27. Available from: http://www.health.gov.au/internet/main/publishing.nsf/Content/cda-pubs-cdi-2003-cdi27suppl-htm-cdi27supl.htm
