Because different patient populations have different health outcomes, simply comparing the number or percentage of some quality measures will not provide an accurate comparison. To make a more accurate comparison, some measures are “adjusted.” This is a common statistical step so that comparison may be similar. However, even adjusting the data cannot guarantee an identical comparison of “apples to apples.”
Adjusting data is a common statistical step to more accurately compare the outcomes among different hospitals. Some hospitals may serve patients that are sicker or poorer than other hospitals. Adjusting the data for the risk of needing care makes it possible to compare performance. For example, a hospital that provides care to older patients will have different outcomes for chronic diseases than a hospital that provides care to a younger group of patients. Therefore, the data for these two hospitals would be age-adjusted to allow comparison. Other common types of adjustments include existing illnesses and sex. In this report, quality information about preventable hospitalizations and readmissions have been adjusted to increase the comparability of data. The preventable hospitalizations have been adjusted for age group and sex. The information for readmissions is adjusted for several factors. Currently, the government agencies responsible for reviewing quality information adjust the readmission data for age, past medical history and other diseases or conditions. However, there is emerging national research that suggest poverty and other community factors increase the likelihood a patient will have an unplanned readmission to the hospital within 30 days of discharge. Based on new research and suggestions, the readmission data in this report have been adjusted for age, past medical history, other diseases or conditions, as well as Medicaid status and neighborhood poverty rate.
Read more about the 30-day risk adjusted readmission rates derived with multi-level models designed to control for patient’s clinical, sociodemographic and community-based risk factors.