Posted by: Jenny Laurello
CDI, Clinical documentation, ICD-10
Guest post by: Wendy Whittington, MD, MMM, Chief Medical Officer, Anthelio Healthcare Solutions, Inc.
A new study in JAMA published last month deserves our attention. The study compares the hospitalization and mortality rates for pneumonia from 2003 to 2009. The authors noted the shift in coding from straight pneumonia as the first diagnosis to sepsis or respiratory failure as the first diagnosis and pneumonia as the second. They also took into consideration the etiology (microbiology) of the pneumonia and the disposition at discharge.  Further, they noted that Present on Admission (POA) was not coded until 2008 and couldn’t be considered in their research. They stated: “These findings have important implications. They suggest that attempts to measure the outcomes of patients with pneumonia by studying only those who receive a principal diagnosis of pneumonia will be biased toward increasingly less severe cases.”
What interests us about this article is the role that “secular” (their term) coding processes played in the change in metrics for pneumonia. Because of clinical documentation improvement (CDI) programs across U.S. hospitals, coding of this disease has apparently become more accurate and sicker patients who actually had sepsis or respiratory failure are documented and coded as such. This makes sense. Physician education programs in CDI have some consistent focuses, with pneumonia, sepsis and respiratory failure certainly among them. One slogan heard in these physician education programs focuses on not using the term “insufficient” if respiratory function has failed: “why be insufficient when you can fail?”
Another principle of physician education attached to CDI processes is Present on Admission (POA). Although this may be viewed as a somewhat artificial concept from CMS, it does help health care providers produce documentation that is more precise, thorough and accurate. The goal of CDI is to make medical records more reflective of the actual condition of the patient. Documentation that is conflicting, incomplete or generic will not help improve the care of patients.
Though the authors of the JAMA article expressed concern about the inability to use these statistics for longitudinal studies of care quality, I would suggest that the goal of CDI programs is to get the documentation right for both patients and providers.. Comparative effectiveness research must be based on the true condition of the patient in order to illuminate further studies of what treatment works and what doesn’t.
We at Anthelio have been supporting CDI programs and the change to ICD-10 coding. With more specific coding in ICD-10, the etiology of pneumonia will be coded with combination codes that will make the needed associations more visible and easier to research using coding data. This scenario is just another example of why coding must be updated to provide an accurate picture of health care as it currently is and not allowed to stagnate for 30 years as with ICD-9 coding.
1. Lindenauer PK. Association of Diagnostic Coding With Trends in Hospitalizations and Mortality of Patients With Pneumonia, 2003-2009. JAMA 2012; 307(13):1405-1413.