Quality Measures

There are 3 types of Home Health Quality Measures based on OASIS-C data: Process Measures, Outcome Measures, and Potentially Avoidable Events.  Additionally,two Home Health utilization measures are based on Medicare Claims data. Process measures have been reported since October 2010 and  "Outcome" and "Potentially Avoidable Event" measures have been reported since June 2011. Claims-based utilization measures were first included on CASPER reports in August 2012 and will be added to Home Health Compare in late 2012/early 2013. This page contains brief descriptions of each type of measure and tables identifying which Home Health Quality Measures are risk adjusted and reported publicly. 

Outcome Measures 

Outcome measures report a change (or lack of change) in patient condition during an episode of care. There are two types of outcomes--utilization outcomes and end-result outcomes. Higher values are preferable to lower values for end result outcomes, while lower values are preferable for utilization outcomes. 

Outcome measures are calculated based on a completed episode of care that begins with admission to a home health agency (or a resumption of care following an inpatient facility stay) and ends with discharge or transfer to inpatient facility. This is different than a Home Health Prospective Payment episode of 60 days. 

Risk Adjustment - All the outcome measures are risk adjusted. The risk adjustment methodology, using a predictive model developed specifically for each measure, compensates for differences in the patient population served by different home health agencies. 

In the Home Health Quality Outcomes Measure Table, located via the link for Tables for Outcome, Potentially Avoidable Event and Process Measures [ZIP 33 KB] in the Downloads section below, the OBQI outcome measures for which the predictive models did not meet the criteria R2 > 0.10 or C-statistic > 0.70 are designated with a pound sign (#) to highlight the fact that they have less robust risk adjustment. For more information about the outcome measures please follow the link in the Downloads section below for the OBQI Manual.

Potentially Avoidable Event Measures

Potentially avoidable events serve as markers for potential problems in care because of their negative nature and relatively low frequency. The potentially avoidable events reported are outcome measures, in the sense that they represent a change in health status between start or resumption of care and discharge or transfer to inpatient facility. All the Potentially Avoidable Event Measures are adjusted for variation in patient characteristics. The predictive model for Development of Urinary Tract Infection did not meet the criteria R2 > 0.10 or C-statistic > 0.70 and so is designated with a pound sign (#) to highlight the fact it has less robust risk adjustment. Please check the table for "Home Health Quality Potentially Avoidable Event Measures" by accessing the link to Tables for Outcome, Potentially Avoidable Event and Process Measures [ZIP 33 KB] in the Downloads section below. 

For more information about the Potentially Avoidable Event Measures please follow the link in the Downloads section below for the OBQM Manual.

Process Measures

Process quality measures evaluate the rate of home health agency use of specific evidence-based processes of care. The OASIS-C process measures focus on high-risk, high-volume, problem-prone areas for home health care. These include measures pertaining to all or most home care patients, such as timeliness of home care admission, immunizations, and use of risk assessment tools for falls, pain, depression, and pressure ulcer development. As well, there are measures for specific diagnoses (heart failure, diabetes, pressure ulcers) and measures of care planning and clinical interventions delivered for patients experiencing certain symptoms (pain, depression).

Process measures are calculated based on a completed episode of care that begins with admission to a home health agency (or a resumption of care following an inpatient facility stay) and ends with discharge or transfer to inpatient facility. This is different than a Home Health Prospective Payment episode of 60 days.

Unlike the outcome, potentially avoidable event, and utilization measures, process measures are not risk adjusted. Risk adjustment is not considered to be necessary for process measures because the processes being measured are appropriate for all patients included in the denominator (patients for whom the measure is not appropriate are excluded). Please check the table for "Home Health Quality Process Measures" by accessing the link to Tables for Outcome, Potentially Avoidable Event and Process Measures [ZIP 33 KB] in the Downloads section below.

 For more information about the process measures please follow the link in the Downloads section below for the PBQI Manual.

Claims-based Utilization Measures

Utilization quality measures evaluate the rate of potential problems in care that are indicated by the utilization of specific services. Utilization measures are based on Medicare claims data. Lower values are preferable to higher values because they indicate fewer negative events for patients at that home health agency.

Utilization measures are calculated based on the first home health claim that starts an episode of care for a patient and include negative events such as hospitalization or emergency department care within a certain time frame.  Detailed specifications for the utilization measures can be found in the Claims Based Utilization Measures Specifications link below under Downloads.

Risk Adjustment – The utilization measures are risk adjusted using a predictive model developed specifically for the measures which takes into account differences in patient health status, as measured by the patient’s previous Medicare claims. The risk adjustment methodology can be found in the Claims Based Utilization Measures Risk Adjustment Methodology link below under Downloads.