OBJECTIVES: To compare results of regression and statistical process control chart analysis for hospital-acquired pressure ulcers (HAPUs).  To determine which analytical models are best interpreted for the clinical microsystem. METHODS: Data were extracted from billing data and patient records at a tertiary-care facility in the US from 2004-2007 for parameters associated with HAPU incidence including: age, gender, medical history, pressure ulcer location, length-of-stay, and pressure ulcer risk assessment with the Braden scale. Primary outcome measures were HAPU incidence rate, days between diagnoses of HAPUs in an inpatient surgical ward, and proportion of completed Braden scales for admitted post-operative patients.  These outcomes were first analyzed in linear probability model and tested for heteroskedasticity, then as a panel data set.  Finally, data were fit to statistical process control charts for a closer look at performance in clinical microsystems. RESULTS: Among 43,844 hospital inpatients, there were 337 total incidences of HAPUs hospital-wide. A probit regression model predicted the correlation of age, gender, and length-of-stay on HAPU incidence (R2 = 0.096).  Panel data analysis determined that for each additional day in the hospital, there was a 0.3% increase in the likelihood of developing a HAPU.  A p-chart of HAPU incidence showed a mean incidence rate of 1.17% that remained in statistical control. Based on the g-chart, the average time between events for the last 25 HAPUs was 13.25 days.  There was a 55-day period between two incidences during the observation period. The p-chart addressing Braden Scale assessments showed that 40.5% of all patients were risk-stratified upon admission. CONCLUSIONS: Standard regression analysis is limited to hospital-wide data, and should not be used to interpret disease outcomes at the level of the clinical microsystem.  Statistical process control charts amplify patient outcomes at the point of care, and are useful for guiding and developing quality improvement interventions.