Patients with Hospital‐Acquired Pressure Ulcers: Inaccurate Documentation and Poor Outcomes

Background:

Pressure ulcers are a significant source of morbidity in hospitalized patients. In 2006 the Centers for Medicare and Medicaid Services (CMS) identified the development of a stage 3 or 4 pressure ulcer while hospitalized as a “never event.” This designation along with the requirement that physicians take part in ulcer documentation prompted increased interest in ulcer prevention and treatment. Yet little is known about the accuracy of pressure ulcer documentation and coding.

Methods:

The University Healthsystem Consortium (UHC) undertook a benchmarking project in 2008 that retrospectively reviewed the charts of 6100 patients at 32 hospitals. The inclusion criteria for all possible cases were patient age ≥ 18 and LOS > 4 days. Patients were excluded if there was an ICD‐9 code for pressure ulcer present on admission (POA) or spinal cord injury. Two study groups were identified: patients with a code for pressure ulcer, and those identified to be at high risk (coded with 1 of 14 designated DRGs, examples of which include septicemia and cardiac arrest). Staff at each hospital reviewed the charts using a standardized instrument.

Results:

Of the 2035 patients coded as having a pressure ulcer that was not POA, only 32.4% actually had a pressure ulcer acquired during the hospitalization. In the high‐risk group of more than 4000 patients, 10.8% had an ulcer on admission, and 14.8% developed an ulcer, but none were coded as having ulcers (Table 1). Chart review showed 96.5% of patients had skin assessments and 80.9% had ulcer risk assessments completed on admission. However, when an ulcer developed, only 77.3% of patients had the ulcer stage documented. Among all patients with acquired pressure ulcers the median length of stay was 26 days. Among hospitalized patients, 68.9% of patients were discharged with the ulcer still present, 15.5% entered hospice care, 20.3% died during the hospitalization, and 19.6% were readmitted to the same institution within 30 days.

Conclusions:

Coding for pressure ulcers appears to be inaccurate and not a reliable source to determine the prevalence of hospital‐acquired pressure ulcers. Compliance with process measures such as skin and risk assessments shows that hospitals attempt to identify and prevent pressure ulcers, but the lack of proper documentation and coding may be a barrier to better identification of patients in need of intensive skin care. Developing a pressure ulcer indicates an extremely poor prognosis.

Author Disclosure:

C. Ritter, none; M. Williams, none; D. Malkenson, none; L. Grogean, none; S. Zarathkiewicz, none; P. Romano, none; E. Streib, none; C. Perumalswami, none; J. Garrett, none.