Scoring Method
In order to provide a valid and reliable comparison of quality performance across providers, The Delta Group utilized a number of statistical methods prior to calculating percentile scores for each quality measure, quality measurement category, and clinical category. Specifically, The Delta Group:
-
Removed statistical outliers at three (3) standard deviations from the mean of the log distribution of charges and length of stay within each MS-DRG terminal grouping (based on severity level, age cohort and CC cluster).
- Evaluated hospital performance for each quality measure within the following quality measurement categories:
- Mortality Overall
- Complications Overall
- Inpatient Quality
- Core Process
- Patient Safety*
- Patient Satisfaction
*Excluded the following patient safety indicators (PSIs) from public reporting due to inconsistent coding practices among hospitals or the prevalence of false positives: complications of anesthesia (PSI 1), death in low mortality DRGs (PSI 2), accidental puncture and laceration (PSI 15), and transfusion reaction (PSI 16).
- Assigned each quality measure to one or more of the following clinical categories:
- Overall Hospital Care
- Overall Medical Care
- Overall Surgical Care
- Cardiac Care
- Major Cardiac Surgery
- Orthopedic Care
- Major Orthopedic Surgery
- Neurological Care
- Major Neuro-Surgery
- General Surgery
- Cancer Care
- Respiratory Care
- Women’s Health (All-Payer Data Only)
- Each hospital’s performance on each measure was tested for its level of statistical significance using a z-value (based on the one-proportion z-test) which was then converted to a z-score to convert all data to a standard normal distribution (z-scores ensure an appropriate statistical distribution and equal weighting where each z-score corresponds to a point on a standard normal distribution characterized by a new mean of zero and a new standard deviation of 1; with an approximate range of -3 and +3 standard deviations).
This statistical routine accomplishes the following:
- Accounts for each hospital’s standard deviation from the mean and the effect of their respective caseload volume (z-value).
- Prior to calculating percentile performance, the z-score converts all underlying distributions (e.g., AMI-1 core measure, AMI-2 core measure, PSI 1, PSI 2, and etc.) to a standard normal distribution so that percentile performance can be accurately identified within and across quality measurement categories.
By incorporating both z-values and z-scores, this approach produces normally-distributed “statistically significant percentiles of performance” for accurately and uniformly assessing the quality of hospital care.
Eligibility Requirement for Composite Scoring and Ratings
Hospitals must receive a composite quality score ≥ 90th percentile with no lower than a "√" rating for any measurement category.
Rating Method
Using Table 1 below, each hospital’s quality rating was determined by assigning its respective percentile score to the appropriate percentile reference range and identifying the associated quality rating (√++ through √--).
Distribution of Quality Ratings
The percentile reference ranges distribute hospital quality ratings as follows*:
- 10% receive the “highest” quality rating (√++)
- 15% receive a “high” quality rating (√+)
- 50% receive an “average” quality rating (√)
- 15% receive a “low” quality rating (√-)
- 10% receive the “lowest” quality rating (√--)
*Prior to the removal of hospitals which do not meet eligibility requirements