New York State 2015 Nursing Home Quality Initiative Results and 2016 Proposals New York State Department of Health March 7, 2016

2 NHQI Status Updates

NHQI State Planning Amendment was submitted to CMS in June 2015 Facility-specific results were released on the Health Commerce System in January 2016 Quintile ranking was released on the Department’s Medicaid Redesign Team (MRT) website in January 2016 Downloadable data was released on Health Data NY in February NHQI Status Updates

NHQI Methodology

NHQI Structure Quality Component: 70 points Percent of Long Stay High Risk Residents With Pressure Ulcers* Percent of Long Stay Residents Who Received the Pneumococcal Vaccine (scored by quintiles) - change from 2014 NHQI Percent of Long Stay Residents Who Received the Seasonal Influenza Vaccine Percent of Long Stay Residents Experiencing One or More Falls with Major Injury Percent of Long Stay Residents Who have Depressive Symptoms Percent of Low Risk Long Stay Residents Who Lose Control of Their Bowels or Bladder Percent of Long Stay Residents Who Lose Too Much Weight* Percent of Long Stay Antipsychotic Use in Persons with Dementia (PQA) - change from 2014 NHQI Percent of Long Stay Residents Who Self-Report Moderate to Severe Pain* Percent of Long Stay Residents Whose Need for Help with Daily Activities Has Increased Percent of Long Stay Residents with a Urinary Tract Infection Percent of Employees Vaccinated for Influenza CMS Five-Star Quality Rating for Staffing (not NYS-calculated) Percent of Contract/Agency Staff Used *denotes risk adjustment by NYS Compliance Component: 20 points NYS Regionally Adjusted Five-Star Quality Rating for Health Inspections Timely Submission of Nursing Home Certified Cost Reports Timely Submission of Employee Influenza Immunization Data Efficiency Component: 10 points Number of Potentially Avoidable Hospitalizations per 10,000 Long Stay Days*

6 Scoring Details - Quality Component Quintile 1: 5 points Quintile 2: 3 points Quintile 3: 1 point Quintiles 4 and 5: 0 points Two measures were awarded 5 or 0 points based on threshold values: 1.Percent of Contract/Agency Staff Used (5 points for a rate of less than 10%) 2.Percent of Employees Vaccinated for Influenza (5 points for a rate of 85% or higher) Nine measures were eligible for improvement points based on the previous year’s quintile 1.Percent of Long Stay High Risk Residents With Pressure Ulcers 2.Percent of Long Stay Residents Experiencing One or More Falls with Major Injury 3.Percent of Long Stay Residents Who have Depressive Symptoms 4.Percent of Low Risk Long Stay Residents Who Lose Control of Their Bowels or Bladder 5.Percent of Long Stay Residents Who Lose Too Much Weight 6.Percent of Long Stay Residents Who Self-Report Moderate to Severe Pain 7.Percent of Long Stay Residents Whose Need for Help with Daily Activities Has Increased 8.Percent of Long Stay Residents with a Urinary Tract Infection 9.Percent of Long Stay Residents who Received the Seasonal Influenza Vaccine 2014 Performance 2015 Performance Quintile (best) If 2014 NHQI performance was in the third quintile, and 2015 NHQI performance was in the second quintile, the facility received 4 points. This is 3 points for attaining the second quintile and 1 point for improvement from the previous year’s third quintile.

7 Scoring Details – Small Sample Size Facility A No small sample size (no reduction in base points) Facility B Small sample size on two quality measures, each worth 5 points (10-point reduction in base points) Sum of points80 Base points10090 Score (points/maximum) Score x 10080%89% If small sample size occurred for a measure, or the measure was unavailable, the maximum attainable score was reduced

8 Small Sample Size in Risk Adjusted Quality Measures Old method CMS Nursing Home Compare reports rates for risk adjusted measures if the total denominator (sum across three quarters) is at least 30 CMS risk adjusts by quarter and averages to create a three-quarter average NHQI follows the same method to create a four-quarter average Modified method If the denominator in any quarter is less than 30, DOH does not consider the quarterly risk adjusted rate to be reliable For nursing homes where only one quarterly denominator is less than 30, DOH will substitute the statewide quarterly average and calculate a four-quarter average For nursing homes where two or more quarterly denominators are less than 30, DOH will suppress the measure in the NHQI and reduce the base

9 Scoring Details – Compliance Component Compliance Component NYS Regionally Adjusted Five-Star Quality Rating for Health Inspections o Used CMS health inspection survey scores as of April 2015 to calculate cut points for each region in the state o Regions include the Metropolitan Area, Western New York, Capital District, and Central New York o Within each region, the top 10% of nursing homes received five stars, the middle 70% received four, three, or two stars, and the bottom 20% received one star o Each nursing home was awarded a Five-Star Quality Rating based on the cut points calculated from the health inspection survey scores within its region o 10 points for 5 stars, 7 points for 4 stars, 4 points for 3 stars, 2 points for 2 stars, 0 points for 1 star Timely Submission of Nursing Home Certified Cost Reports – 5 points Timely Submission of Employee Flu Immunization Data – 5 points Efficiency Component Potentially Avoidable Hospitalizations o Quintile 1: 10 points o Quintile 2: 8 points o Quintile 3: 6 points o Quintile 4: 2 points o Quintile 5: 0 points

10 Scoring Details - Efficiency Component Potentially Avoidable Hospitalizations Quintile 1: 10 points Quintile 2: 8 points Quintile 3: 6 points Quintile 4: 2 points Quintile 5: 0 points Old method One Potentially Avoidable Hospitalizations (PAH) rate is calculated from the measurement year New method PAH rate will be calculated for each quarter to align with the other quality measures Quarterly rates will be risk adjusted and averaged to create an annual average PAH rate (same method for other risk adjusted measures) Denominator will be the number of days contributed from each long stay resident

NHQI Results

NHQI Measure Statistics (2014 unshaded for comparison) 12 Measure P100 (max/min) P80P60P40P20 P0 (max/min) Statewide Average Overall score Percent of long stay residents who received the pneumococcal vaccine Percent of long stay residents who received the seasonal influenza vaccine Percent of long stay high risk residents with pressure ulcers Percent of long stay low risk residents who lose control of their bowel or bladder Percent of long stay residents experiencing one or more falls with major injury Percent of long stay residents who have depressive symptoms Percent of long stay residents who lose too much weight Percent of long stay residents with dementia who received an antipsychotic medication Percent of long stay residents who self-report moderate to severe pain Percent of long stay residents whose need for help with daily activities has increased Percent of long stay residents with a urinary tract infection Number of potentially avoidable hospitalizations per 10,000 long stay days Higher rate is better Lower rate is better 1 st Quintile 3 rd Quintile 2 nd Quintile 4 th Quintile 5 th Quintile Percent of long stay residents who received the seasonal influenza vaccine 78% or less: 5 th quintile Greater than 78% but less than or equal to 83%: 4 th quintile Greater than 83% but less than or equal to 87%: 3 rd quintile Greater than 87% but less than or equal to 91%: 2 nd quintile Greater than 91%: 1 st quintile Percent of long stay high risk residents with pressure ulcers 4.8% or less: 1st quintile Less than or equal to 6.3% but greater than 4.8%: 2 nd quintile Less than or equal to 7.6% but greater than 6.3%: 3 rd quintile Less than or equal to 9.5% but greater than 7.6%: 4 th quintile Greater than 9.5%: 5 th quintile Interpreting Quintiles

13 Quality Component – Employee flu vaccine Statewide employee influenza vaccination average remains at 86% in 2013 and 2014 Measurement year NHQI year Statewide Average % % %

14 Quality Component – Improvement Results Facilities received one point for improvement if the 2015 NHQI quintile for a measure was an improvement from the 2014 NHQI quintile 93% of facilities received at least one improvement point (compared to 94% in 2014 NHQI) Number of Improved Quality Measures Percent Facilities 2014 NHQI 2015 NHQI or more107 Total9493

15 Quality Component – Resident Immunization Measures Percent of long stay residents who received the pneumococcal vaccine 2014 NHQI2015 NHQI STW85% 84 % Quintile 1 (MAX) 100 % Quintile 2 93 % Quintile 3 89 % 88 % Quintile 4 85 % 83 % Quintile 5 77 % 75 % MIN 17 % 33 % Percent of long stay residents who received the seasonal influenza vaccine 2014 NHQI2015 NHQI STW 83 % 84 % Quintile 1 (MAX) 100 % Quintile 2 91 % Quintile 3 87 % Quintile 4 83 % Quintile 5 77 % 78 % MIN 26 % 50 % 2015 NHQI statewide average is less than 85% Using 85% threshold, 56% of nursing home receive points Using quintiles, 58% receive points Decision for 2015 NHQI: use quintile method to allow more nursing homes to receive points; measure is not eligible for improvement 2015 NHQI statewide average is less than 85% (same as previous years) Decision for 2015 NHQI: continue to use quintile method; measure is eligible for improvement from 2014 NHQI

16 Quality Component – Pharmacy Quality Alliance Antipsychotic Measure CMS Long Stay Antipsychotic Use measure includes residents for whom antipsychotic medication may be appropriate o Does not exclude residents with Manic Depression or Bipolar Disease, but NYS added this exclusion in past NHQIs Pharmacy Quality Alliance (PQA) measure – Antipsychotic Use in Persons with Dementia (long stay) Focuses on residents with dementia who have a history of receiving an antipsychotic (FDA black box warning) Denominator consists of residents with: o Alzheimer’s Disease o Non-Alzheimer Dementia o Moderately or severely impaired cognitive functioning o Brief Interview for Mental Status (BIMS) score indicating cognitive impairment Excludes residents with the diagnoses of Huntington’s Disease, Tourette’s Syndrome, Manic Depression or Bipolar Disease, and Schizophrenia Residents qualify for the numerator if on the last two MDS assessments, the total number of days they received an antipsychotic was 12 or more (look back period of 7 days on each assessment)

17 PQA and CMS Antipsychotic Measure Statistics Measure Measurement year NHQI year Quintile 1Quintile 2Quintile 3Quintile 4Quintile 5 Statewide Average CMS %14%18%24%53%17% PQA %13%17%22%51%16% PQA %12%15%20%51%14% PQA and CMS Quintile Cut Points

18 CMS vs. PQA Antipsychotic Measure Quintile Cut Points 2013 Measurement Year

19 PQA Antipsychotic Measure Quintile Cut Points 2013 and 2014 Measurement Years

20 Compliance Component and Deficiencies Compliance NYS Regionally Adjusted Five-Star Quality Rating for Health Inspections o Used CMS health inspection survey scores as of April, 2015 to calculate cut points for each region in the state o Regions include the Metropolitan Area, Western New York, Capital District, and Central New York o Within each region, the top 10% of nursing homes received five stars, the middle 70% received four, three, or two stars, and the bottom 20% received one star o Each nursing home was awarded a Five-Star Quality Rating based on the cut points calculated from the health inspection survey scores within its region o 10 points for 5 stars, 7 points for 4 stars, 4 points for 3 stars, 2 points for 2 stars, 0 points for 1 star Timely Submission of Nursing Home Certified Cost Reports – 5 points o 97.3% (N=578) of facilities submitted by the designated deadlines (compared to 95% in 2014 NHQI) Timely Submission of Employee Flu Immunization Data – 5 points o 97.5% (N=579) of facilities submitted by the May 1, 2015 deadline (compared to 97% in 2014 NHQI) Deficiencies (July 1, June 30, 2015) 3% (N=16) of facilities received a J, K, or L deficiency, compared to 6% (N=34) in 2014 NHQI NHQI year Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Total facilities with a deficiency Total facilities (<1)3 (<1)7 (1)9 (1)13 (2)34 (6) (<1)4 (<1)7 (1)16 (3)594 Number (%) of facilities with a J, K, or L deficiency in 2014 NHQI compared to 2015, by Quintile

21 Benchmarking Staffing Measures

22 New Staffing Measures Two new staffing measures were reported for feedback with the 2015 NHQI results 1.Rate of staffing hours per day o Using cost report and MDS data o Intent to replace the current CMS Five-Star Quality Rating for Staffing 2.Percent of staff turnover o Using cost report data o Additional staffing measure Measures were reported for benchmarking purposes only Measures were not incorporated into the scoring of the 2015 NHQI

23 Rate of Staffing Hours Per Day CMS staffing measure CMS 5-Star for Staffing based on 2 week snapshot of facility prior to on-site survey Hours and days reported from form CMS-671 and CMS-672 Hours expected computed from the distribution of RUG categories within facility at end of quarter closest to survey, multiplied by the CMS Time Staff Measurement Study Hours reported o Full time hours worked for RNs, LPNs, and Aides: Schedule 5A o Contract hours paid for RNs, LPNs, and Aides: Schedule O – Quality Days reported from Bed Capacity Patient Days schedule Hours and days expected are computed by counting resident days at MDS 3.0-generated RUGIII category, and utilizing CMS Time Staff Measurement Study (Hours reported from CMS-671 / # of residents from CMS-672) / 14 days [(RUG distribution of quarter closest to survey*hours from CMS time study) / # of residents from CMS-672] / 14 days (Annual hours reported on cost report / Annual days reported on cost report) (Annual hours expected from MDS RUG distribution / Annual days expected from MDS) NYS staffing measure Statewide reported average

24 Rate of Staffing Hours Per Day CMS staffing data from Five-Star Provider Ratings, as of April 2015 o National average: 4.0 hours per resident/day o NY average: 3.9 hours per resident/day P0 (min) P20P40P60P80 P100 (max) NY Staffing Quintile Cut Points, 2014 Change Number and percent of facilities (N=549) Declined by 3+ rankings152.7% Declined by 2 rankings295.3% Declined by 1 ranking7313.3% No change19135% Improved by 1 ranking % Improved by 2 rankings519.3% Improved by 3+ rankings244.4% CMS Five-Star Quality Rating for StaffingNY Rate of Nursing Hours per Day Nursing home CMS Reported hours/resident/day CMS Expected hours/resident/day CMS Adjusted hours/resident/day CMS Star Rating NY Reported hours/resident/day (cost report) NY Expected hours/resident/day (MDS) NY Adjusted hours/resident/day NY Quintile A star (lowest) (highest) B stars (second highest) (lowest) 78.9% no change or improvement Change between CMS Five-Star Quality Rating for Staffing and NY Staffing Quintile NY 2014 nursing home cost report data o NY average: 3.5 hours per resident/day

25 Rate of Staffing Hours Per Day – Example Format Data FieldValue 1. Cost Report Data Total reported staffing hours (full-time and contract)200,000 Total reported bed capacity patient days60,000 Reported staffing hours per day MDS Data Total expected staffing hours (RUG-adjusted)210,000 Total expected patient days (RUG-adjusted)60,500 Expected staffing hours per day (RUG-adjusted) Statewide Average Statewide reported staffing hours per day Result Adjusted staffing hours per day3.3 Quintile (hypothetical)2 In this example, the facility’s reported staffing hours per day is 3.3 Based on the facility’s RUG distribution from the MDS, the expected staffing hours per day is 3.5 The adjusted staffing hours per day is hours per day Hypothetically, the facility would qualify for the second quintile in this measure 3.3 reported 3.5 expected 3.5 statewide average

26 Rate of Staffing Hours per Day – Discussion Feedback o No feedback was received regarding the benchmarking measure Cost report limitations o Nurses with management, supervision, or administrative duties may also provide direct care to patients, but may not be included in the staffing measure, depending on how the cost report is completed Benchmarking o Measure is difficult to benchmark to CMS Five-Star Quality Rating for Staffing due to different data sources and measurement periods

27 Percent of Staff Turnover Adapted methodology from the Advancing Excellence staff turnover calculation Changes to the 2014 cost report allowed for distinction between full-time staff and per diem staff o Schedule O – Quality: fields were added to capture turnover data for per diem RNs, LPNs, and Aides staff Benchmarking turnover rates were reported with the 2015 NHQI results Date Total RNs, LPNs, and Aides End Quarter 1160 End Quarter 2150 End Quarter 3155 End Quarter 4160 TOTAL625 Example Calculation 1.Sum of the quarter-end total number of non-contract, non-per diem RN, LPN, and Aide staff = Average number of RN, LPN, and Aide staff members per quarter (625 / 4 quarters) = Assume total number of RN, LPN, and Aide staff terminated throughout the year = 24 4.Number of RN, LPN, and Aide staff members terminated = average terminated / average number of staff members = 15.4% annual staff turnover rate

28 Percent of Staff Turnover – Discussion Feedback o No feedback was received regarding the benchmarking measure Unique vs. non-unique numerator and denominator o Number of unique employees terminated in the year divided by a non-unique annual average Cost report limitations o Some facilities reported a greater number of per diem employees than full-time and per diem employees combined, resulting in a negative percent of staff turnover Example o Questionable data validity (staff turnover values ranged from -106% to 1,231%) o Unable to determine valid quintile cut points Year Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Statewide %23%36%57%1,231%35.4% Number of full-time and per diem employees terminated, combined (Schedule P)10 Number of per diem employees terminated (Schedule O)15 10 combined – 15 per diem-5 full-time employees terminated

29 Percent of Staff Turnover – Discussion Seasonal staff o Depending on how the cost report is completed, the measure may or may not capture seasonal employees and students, which would affect staff turnover rate Defining per diem o No definition of per diem staff for nursing home employees Alternative measure exploration o Explored a staff retention measure as an alternative to staff turnover o Same cost report issues remain (per diem and seasonal staff)

30 Staffing Measures – Discussion Quality Component will continue to have two staffing measures o Percent of contract/agency staff used o Rate of staffing hours per day (new proposed NYS measure to replace CMS Five-Star Quality Rating for Staffing) Importance of staff-type measures in quality component Impact of adding a third staffing measure to quality component (turnover or retention) Other priority areas to explore o Efficiency measures

NHQI and Future Items

NHQI Structure Quality Component: 70 points Percent of Long Stay High Risk Residents With Pressure Ulcers* Percent of Long Stay Residents Who Received the Pneumococcal Vaccine Percent of Long Stay Residents Who Received the Seasonal Influenza Vaccine Percent of Long Stay Residents Experiencing One or More Falls with Major Injury Percent of Long Stay Residents Who have Depressive Symptoms Percent of Low Risk Long Stay Residents Who Lose Control of Their Bowels or Bladder Percent of Long Stay Residents Who Lose Too Much Weight* Percent of Long Stay Antipsychotic Use in Persons with Dementia (PQA) Percent of Long Stay Residents Who Self-Report Moderate to Severe Pain* Percent of Long Stay Residents Whose Need for Help with Daily Activities Has Increased Percent of Long Stay Residents with a Urinary Tract Infection Percent of Employees Vaccinated for Influenza Rate of Staffing Hours per Day (new proposed NYS measure) Percent of Contract/Agency Staff Used *denotes risk adjustment by NYS Compliance Component: 20 points NYS Regionally Adjusted Five-Star Quality Rating for Health Inspections Timely Submission of Nursing Home Certified Cost Reports Timely Submission of Employee Influenza Immunization Data Efficiency Component: 10 points Number of Potentially Avoidable Hospitalizations per 10,000 Long Stay Days*

33 Recognizing Continued Top Performance 33 nursing homes qualified for the top quintile across the 2013, 2014, and 2015 NHQIs An “Honor Roll” would be created to recognize nursing homes that attained the top quintile across three years of the NHQI Updated with each release of the NHQI to reflect a three-year rolling list List would be included in the Quintile Ranking spreadsheet, and would be visible from the NHQI homepage on the Department’s MRT website

34 MDS Section S Dental Care, effective October 2014 S7000. Dental Care 1.Routine dental care since last assessment – planned 2.Emergent dental care since last assessment – unplanned 9. None of the above Section S items submitted to CMS o Capturing residents who are receiving comfort care o Capturing hospitalizations that are the result of a family member’s request o If approved by CMS, effective date would be October 2017

35 Future Measures Resident satisfaction Dental health Efficiency measures

36 Questions/Comments Methodology Office of Quality and Patient Safety (518) Rate Adjustments Division of Finance and Rate Setting