The impact of miscoding of community -acquired pneumonia in a UK district general hospital

Background: The incidence of Community-Acquired Pneumonia (CAP) in the United Kingdom is 5-11/1000. 1 Misdiagnosis is common. 2 Coding information, following episodes of inpatient care, is derived from diagnoses recorded in clinical notes. There is little published data on financial implications of incorrect coding in CAP. Method: We conducted a retrospective review of records incorrectly coded as CAP between March 2012 and March 2013, and assessed whether this had an impact on length of stay and income. Patients with clear radiographic evidence (CXR or CT) of pneumonia were excluded along with those who actually had HAP or were immunocompromised.We sought to determine if miscoding of CAP (ICD-10) would negatively impact on income streams and length of stay. We analyzed length of stay and Healthcare Resource Group (HRG) codes for each miscoded episode of care. Results: 602 patients were coded as pneumonia. 233 (38.7%) were accurately coded .195 (32.3%) were miscoded as CAP (non-CAP). Of the latter 32 were excluded due to lack of data, leaving 163 patients whose records were then examined. Alternative diagnoses included; Infective exacerbation of COPD (9%), non-infective exacerbation COPD (9%), lower respiratory tract infection (45%), heart failure (10%), HAP (12%) and other (17%). Median age for CAP was 78 yrs (range: 31-102) and non-CAP was 83 yrs (36-99). Mean length of stay was 10 days for CAP and 12 days for non-CAP (P=0.14). Mean income for each episode of care was € 1124 for CAP and € 1288 for non-CAP (P <0.0001). Conclusions: Our data show that a misdiagnosis and subsequent miscoding of CAP may lead to an apparent increase in length of stay and can lead to a real loss of income .