No abstract

No abstract

No abstract

Statistics regarding aspects of hospital inpatient care are readily available in the public domain. This data is used by policy makers, healthcare commissioners and patients, to compare healthcare providers and inform decision-making. However, by convention these statistics are expressed in the form of the arithmetic mean, which is not an optimal tool for comparing healthcare providers. The authors propose that when comparing lengths of inpatient stay following hospital admissions of elective joint replacement surgery, the geometric mean and mode should be used. These measures are more meaningful to patients, and less sensitive to long stay outliers, which some specialist hospitals are predisposed to due to complexity of case mix, as well as for geographic and socioeconomic reasons. We conducted a comparative cohort study, reviewing prospectively collected length of stay data, for a central London teaching hospital and a Home Counties district general hospital. Our results support the use of the geometric mean and mode over the measures currently used.