FHA Quality Initiatives: Understanding the Star Ratings
Background • FHA Board Goal to Improve Care in Florida • FHA Quality and Patient Safety Committee More Florida Hospitals as 4 or 5 Stars
Star Ratings 45% Florida National 65 40% 35% 1,187 1,155 30% 25% 753 33 34 32 20% 15% 337 10% 260 5% 5 0% One Star Two Stars Three Stars Four Stars Five Stars
Steps to Increasing 4 and 5 Star Hospitals • Increase understanding of the Star Ratings • Offer improvement resources, tools and support • Identify high performing hospitals and their strategies
Hospital STAR Ratings July 2018 Presented by: Kimberly Rask, MD PhD Chief Data Officer Kimberly.Rask@Allianthealth.org 7/13/2018 5
Key Objectives ► Purpose of star ratings ► Measures that are included in star ratings ► How those measures are translated into a rating ► Impact of recent changes in methodology ► Potential future changes ► Strategies for high performance
Purpose of Star ratings ► Over 100 measures on Hospital Compare ► Complaints by consumer groups that it is intimidating and difficult to compare hospitals ► User-friendly format that lets consumers gauge a summary rating across multiple dimensions of quality
What is included in Star rating ► Current measures on Hospital Compare ► Will evolve as measures are added or removed ► Different measures reported by different hospitals ► For the same measure, each hospital may report a different number of cases ► Some metrics are updated quarterly, some annually
Translating Hospital Compare Measures into a Star Rating
Step 1: Picking the measures ► Start with all measures on Hospital Compare ► Delete any measures that are suspended, retired or delayed ► Exclude any measures not being currently publicly reported ► Exclude measures reported by <100 hospitals ► Exclude most structural measures ► Exclude non-directional measures
Step 2: Grouping the Measures Measure Group Individual Measures es Mortality AMI, CABG, COPD, HF, PN, Stroke; PSI- surgical complications Readmission EDAC for AMI, HF, PN; READM for CABG, COPD, Hip-Knee, Stroke, hospital-wide; visit after outpatient colonoscopy Safety of Care CLABSI, CAUTI, SSI-colon, SSI-Hysterectomy, MRSA, c diff, complication following Hip-Knee, PSI composite Patient Cleanliness, nurse and MD communication, Experience responsiveness, medications, discharge, overall rating, quietness, transition measure, willingness to recommend Efficient Use of MRI low back pain, Thorax and Abd CT with contrast, pre- Medical Imaging op cardiac imaging, simultaneous sinus and brain CT Timeliness of ED times, fibrinolysis times, ECG time, time to transfer for Care cardiac intervention, time to evaluation, fracture med time Effectiveness of Influenza, ASA, stroke scan results, polyp surveillance, Care elective delivery, sepsis, VTE, XRT for bone mets
Steps 3: Calculating a Group Score Standardize individual measures so on same 1. scale Z-score: higher value always better • Group into categories similar to HVBP program 2. Calculate group scores from latent variable 3. models (LVM) Measures that are more consistent with each other and • measures that have larger denominators have more influence The influence of individual measures will change with every • cycle Modification added adaptive quadrature 4. enhancement to the model
Step 4: Generate Summary Scores ► Policy-based weighting for measure groups – Measure importance – Consistency – Policy priorities – Stakeholder input ► Re-distribute group weights if hospital does not report any in that group ► Generate summary scores as a weighted average of group scores
Weighting the Measures
Step 5: Apply reporting threshold ► Previously this happened AFTER the Star ratings were calculated ► Now only calculate Star ratings for hospitals for whom they will be reported – Must have 3 measure groups, one of which is an outcomes group, with 3 measures each – 80% of hospitals meet threshold
Translating Hospital Compare Measures into a Star Rating
Step 6: Assign Star Ratings • Apply k- clustering with multiple iterations to assign hospitals to one of five Star Ratings categories • Hospitals are more similar to others in their group than to hospitals in other groups St Star ar Description Cluster of hospitals with highest summary scores Cluster with higher than average summary scores Cluster of hospitals with average summary scores Cluster with below average summary scores Cluster of hospitals with lowest summary scores
July Update to Hospital Compare Delayed ► Concerns about the weighting process 1. Number of measures reported can affect the rating ? Bias toward specialty hospitals and against major teaching • hospitals 2. Differential weighting intended to “separate” hospitals in the ranking process means that an individual measure may be much more heavily weighted than other measures in the same group Changes from quarter to quarter • Example of PSI-90 and complications from Hip-Knee following • transition to ICD-10
Hospital Specific Reports ► CMS shares Overall Hospital Quality Star Ratings Hospital Specific Reports (HSRs) approximately 2 month prior to public reporting ► Hospitals receive a QualityNet notification and have a 30 day review period ► Reports include summary scores, performance category and standardized individual measure scores
Overall Hospital Quality Star Rating Hospital-Specific Report (HSR) Table 1: Overall Hospital Quality Star Rating Results for Your Hospital and the Nation HOSPITAL NAME Results corresponding with data for July 2018 public reporting on Hospital Compare Overall Hospital Rating Your Hospital's Results National Average Results Star Rating [a] ** (2 out of 5 stars) *** (3 out of 5 stars) Hospital Summary Score [b] -0.29 -0.02 Hospital Summary Score Confidence Interval - Lower -0.64 -- Limit [c] Hospital Summary Score Confidence Interval - Upper 0.06 -- Limit [c] [a] A star rating is categorized as one to five whole stars or "N/A". A greater number of stars indicates better performance. The National Average column shows the average star rating across the nation. [b] A summary score is used to determine the star rating category and is calculated from each hospital's measure group scores shown in Table 2. A higher summary score indicates better performance. [c] The 95% confidence interval for your hospital's summary score. The lower confidence limit and upper confidence limit are provided, with a 95% confidence that your performance falls within this range for summary scores.
Table 2: Measure Group Score Results and Weights Table 2: Measure Group Score Results and Weights for the Overall Hospital Quality Star Rating HOSPITAL NAME Results corresponding with data for July 2018 public reporting on Hospital Compare Number of Potential Number of Your Hospital's Standard Group Confidence Measure Group National Group Performance Measure Group Measures within Each Measures for Your Measure Group Measure Group Intervals - Lower Limit, Score [d] Score [f] Category [g] Group [a] Hospital [b] Weight [c] Weight Upper Limit [e] Same as the national Mortality 7 4 22.0% 22.0% -0.16 -1.51, 1.20 0.0004 average Same as the national Readmission 9 5 22.0% 22.0% -0.29 -0.57, -0.02 -0.06 average Same as the national Safety of Care 8 5 22.0% 22.0% 0.16 -0.28, 0.60 -0.04 average Below the national Patient Experience 10 10 22.0% 22.0% -0.98 -1.61, -0.36 -0.0004 average Efficient Use of Same as the national 5 3 4.0% 4.0% 0.64 -0.36, 1.63 0.003 Medical Imaging average Below the national Timeliness of Care 7 6 4.0% 4.0% -0.75 -1.14, -0.35 -0.02 average Same as the national Effectiveness of Care 11 9 4.0% 4.0% -0.14 -1.02, 0.73 0.03 average
Table 3: Individual Measure Scores Table 3: Individual Measure Score Results for the Overall Hospital Quality Star Rating HOSPIITAL NAME Results corresponding with data for July 2018 public reporting on Hospital Compare Your Hospital's Measure Measure’s Measure's Standard Your Hospital's Measure Group [a] Measure ID [b] Measure Name [c] Result on Hospital National Mean of Deviation Across Standardized Compare [d] Scores [e] Hospitals [f] Measure Score [g] Mortality MORT-30-AMI Acute Myocardial Infarction (AMI) 30-Day Mortality Rate 13.0% 13.2% 0.01 0.17 MORT-30-CABG Coronary Artery Bypass Graft (CABG) 30-Day Mortality Mortality N/A 3.2% 0.01 -- Rate MORT-30-COPD Chronic Obstructive Pulmonary Disease (COPD) 30-Day Mortality 8.0% 8.4% 0.01 0.36 Mortality Rate Mortality MORT-30-HF Heart Failure (HF) 30-Day Mortality Rate 10.1% 11.8% 0.02 1.02 Mortality MORT-30-PN Pneumonia (PN) 30-Day Mortality Rate 19.1% 15.9% 0.02 -1.64 Mortality MORT-30-STK Acute Ischemic Stroke (STK) 30-Day Mortality Rate N/A 14.3% 0.02 -- PSI-4-SURG- Death Rate Among Surgical Inpatients with Serious Mortality N/A 161.78 17.01 -- COMP Treatable Complications
Recommend
More recommend
