Study 1: “8 Company”-- MA (HMO) Plan Data and FFS 5% File in Same Local Areas.

Study 1: “8 Company”-- MA (HMO) Plan Data and FFS 5% File in Same Local Areas

Inpatient Days

Inpatient Admissions

E.R. Visits Outpatient Visits

Office Visits Re- Admissions

Avoidable Admissions

Me-dian

1 2 3 4 5 6 7-70%

-50%

-30%

-10%

10%

30%

50%

70%

90%

Eight-Company Results: Percentage Difference in Utilization Rates, MA Rate vs. FFS Rate (per risk score value)

Study 2: Preliminary MA vs. FFS Based on AHRQ (HCUP) Admissions, 2006

Inpatient Days

Same Quarter Re-Admissions (Same DRG, Any Hospital)

13 Potentially Avoidable

Admissions

MA Rate vs. FFS Rate (per risk score value, patients with admissions)California – All Hospitals (acute care admissions) All Patients -30% -15% -6% Diabetes Patients -35% -21% -10% Heart Disease Patients -30% -14% -5%

Nevada – All Hospitals (acute care admissions) All Patients -23% -33% -6% Diabetes Patients -25% -32% -3% Heart Disease Patients -21% -36% -7%Source: ARHD SID (H-CUP), Medicaid included in FFS. http://www.ahipresearch.org/pdfs/CAvsNV.pdf

http://www.ahipresearch.org/pdfs/CAvsNV.pdf

Study 3: Added Two Large Multi-State For Profit MA HMO Plans (Total Plan/Areas = 18)

Inpatient Days

Inpatient Admissions ER Visits Outpatient

Visits Office Visits Same-DRG Readmissions

Avoidable Admissions

Mean -20% -11% -24% 2% 25% -39% -10%High -3% 2% 11% 54% 90% -12% 18%Low -36% -31% -45% -51% -7% -67% -37%Median -21% -12% -25% 0% 9% -42% -8%Areas 18 18 18 11 11 18 18

Source: Working Paper: Comparisons of Utilization in Two Large Multi-State Medicare Advantage HMOs and Medicare Fee-for-Service in the Same Service Areas (December 11, 2009) http://www.ahipresearch.org/pdfs/MAvsFFS-CO9and10.pdf

http://www.ahipresearch.org/pdfs/MAvsFFS-CO9and10.pdf

Inpatient Admissions

Inpatient Days

Same Quarter, Any DRG Readmissions*

Same Quarter,Same DRG Readmissions*

13 Potentially Avoidable Admissions

MA Rate vs. FFS Rate (Patients with an at Least One Admission, Per Risk Score** Value)

California (2006) -6% -30% -24% -20% -8%

California (2007) -6% -28% -24% -20% -8%

Nevada (2006) -2% -24% -23% -29% -9%

Nevada (2007) 2% -17% -21% -21% -8%

Washington (2006) 0% -7% -20% -15% -9%

Texas (2007) 2% -5% -17% -11% -10%

North Carolina (2007) 1% 4% -16% -20% -6%

Pennsylvania (2007) 3% 4% 0% -3% 4%

Hawaii (2007) -1% -7% -3% -13% 1%

Arkansas (2007) -1% -12% -34% -40% -19%

Arizona (2007) 2% -1% 4% 8% 6%

Source: AHIP, based on analysis of state hospital discharge public use datasets with encrypted “person” identifiers, provided by AHRQ’s HCUP project (CA, NV, WA, HI, NC, AR, AZ) and by states directly (PA, TX). http://www.ahipresearch.org/pdfs/9State-Readmits.pdf Note: Excludes patients with an admission listing an out-of-state address.*Excludes transfer cases.** Risk scores for FFS and MA enrollees based on age/sex and HCC relative cost values used in Medicare risk adjustment for beneficiaries living in the community, but do not include disease interactive factors, or factors related to disability or institutional status. Person-based risk scores are based on inpatient hospital diagnoses for all admissions, but do not include diagnosis information from other health care services, such as hospital outpatient or physician office visits.

Study 4: Percentage Difference in Risk-Adjusted Utilization Rates, Persons with Admissions, Medicare Advantage vs. FFS

Study 4: Preliminary Assessment of Issues and Concerns with State Hospital Discharge Datasets for MA vs. FFS ComparisonsState

Year(s) AnalyzedCA

2006-7NV

2006-7WA

2006TX

2007NC

2007PA

2007HI

2007AR

2007AZ

2007

A. Person ID Does Not Identify Same Person *

B. Same Person Might Have Multiple Person IDs X

C. Snowbirds and Out-of-State Residents/Patients X X X X X

D. Possible Missing Data X X

E. Identification of MA Enrollees X X X X X X

F. Unusual Risk Scores/ Diagnosis Codes X

G. Sub-state Composition of Enrollment X X

H. Relative Small Number or Share of MA Enrollees X X X X

I. Cannot Exclude Long-Term Hospitals (Psych., Rehab) X X

Overall Confidence in the StateComparisons of MA and FFS High Medium Medium Medium Some Some Not Much Not Much Not Much

Source: Based on analysis of state hospital discharge public use datasets with encrypted “person” identifiers, provided by AHRQ’s H-CUP project (CA, NV, WA, HI, NC, AR, AZ) and by states directly (PA, TX), and FFS 5 percent claims samples in those states.Notes: Excludes patients with admissions listed as having out-of-state addresses.*2007 data from Washington (not used) had incorrect Person IDs and was returned to AHRQ.

Study 4: Example of Data Checking: Tests for Validity of (encrypted) Person IDs

CA NV WA TX HI NC PA AR AZYear Analyzed 2006 2006 2006 2007 2007 2007 2007 2007 2007

Percentages of Patients with Certain Numbers of Admissions1 Admission

FFS 5% Sample 64.3% 61.8% 67.2% 59.5% 69.4% 63.7% 60.0% 58.5% 63.6%FFS H-CUP 64.7% 67.1% 70.6% 63.7% 71.5% 69.3% 62.1% 59.0% 76.3%Percentage Difference 1% 9% 5% 7% 3% 9% 4% 1% 20%

2 AdmissionsFFS 5% Sample 20.9% 21.9% 20.8% 22.4% 20.5% 21.4% 22.4% 23.7% 21.4%FFS H-CUP 21.0% 20.0% 19.1% 21.5% 18.4% 19.3% 22.0% 22.9% 15.9%Percentage Difference 0% -9% -8% -4% -10% -10% -2% -3% -26%

3 AdmissionsFFS 5% Sample 7.9% 8.0% 7.1% 9.1% 5.4% 8.1% 9.2% 8.7% 7.8%FFS H-CUP 7.7% 7.3% 6.3% 8.1% 6.1% 6.7% 8.6% 9.1% 4.7%Percentage Difference -3% -9% -12% -11% 13% -18% -6% 4% -39%

Overall Confidence in the Validity of the Person IDs High Medium Medium Medium Medium Medium Medium High Not Much

Source: Based on analysis of state hospital discharge public use datasets with encrypted “person” identifiers, provided by AHRQ’s H-CUP project (CA, NV, WA, HI, NC, AR, AZ) and by states directly (PA, TX), and FFS 5 percent claims samples in those states.Notes: Excludes patients with admissions listed as having out-of-state addresses.*2007 data from Washington (not used) had incorrect Person IDs and was returned to ARHQ.

Study 4: Sub-state Geographic Composition, Local ComparisonsReadmission Rate (per admission, unadjusted)

FFS MA Percentage Difference

1 Pittsburgh Area 23.1% 20.1% -13% 2 Erie and Northwest 19.5% 17.8% -9% 3 South Central 19.7% 17.2% -13% 4 North Central 17.3% 16.1% -8% 5 Harrisburg/Lancaster/York 17.6% 17.5% 0% 6 Scranton and Poconos 18.2% 14.2% -22% 7 Allentown and East Central 19.3% 18.1% -6% 8 Philly Suburbs 20.7% 19.2% -7% 9 Philadelphia City 22.6% 21.3% -6%Total (Statewide) 20.1% 19.5% -3%Source: AHIP estimates, based on state hospital discharge datasets provided by the state of Pennsylvania.

Study 4: Sub-state Geographic Composition, PA 2007 FFS MA

Admissions Readmissions Admissions Readmissions

Pennsylvania Hospital Regions

1 Pittsburgh Area 68,334 15,779 64,642 13,004

2 Erie and Northwest 31,694 6,184 7,824 1,393

3 South Central 13,041 2,569 7,996 1,375

4 North Central 17,775 3,079 6,230 1,005

5 Harrisburg/Lancaster 56,935 10,005 5,538 971

6 Scranton and Poconos 33,613 6,113 2,790 396

7 Allentown and E Central 45,042 8,697 6,198 1,122

8 Philly Suburbs 57,517 11,905 33,697 6,465

9 Philadelphia City 32,259 7,290 30,209 6,432

Total (Statewide) 356,210 81,621 165,124 32,163

Source: AHIP estimates, based on state hospital discharge datasets provided by the state of Pennsylvania.

PFFS Enrollees Per County, All Counties within a 100-Mile Radius of Philadelphia

Per Hospitalization* Per Patient with At Least One Admission** Per Estimated Enrollee**

MA Rate vs. FFS Rate (Per Hospitalization, with DRG-Based Risk Adjustment**)

California (2006) -18% -26% -40%

California (2007) -18% -26% -38%

California (2008) -17% -27% -39%

Nevada (2006) -17% -23% -29%

Nevada (2007) -18% -22% -28%

Washington (2006) -17% -18% -43%

Texas (2007) -17% -17% -22%

North Carolina (2007) -12% -12% -26%

Source: AHIP, based on analysis of state hospital discharge public use datasets with encrypted “person” identifiers, provided by AHRQ’s HCUP project (CA, NV, WA, NC) and by states directly (TX). http://www.ahipresearch.org/pdfs/9State-Readmits.pdf Note: Excludes patients with an admission listing an out-of-state address.*DRG-Based risk adjustment. Risk measure based on an index of the likelihood of admissions for DRGs that are associated with higher or lower than average rates of readmissions (any DRG), using the 2006-2007 FFS 5 percent sample file as a benchmark (DRG version 24) and the 2008 FFS 5 percent sample file (DRG version 25).**CMS-Style risk adjustment. Risk scores for FFS and MA enrollees based on age/sex and HCC relative cost values used in Medicare risk adjustment for beneficiaries living in the community, but do not include disease interactive factors, or factors related to disability or institutional status. Person-based risk scores are based on inpatient hospital diagnoses for all admissions, but do not include diagnosis information from other health care services, such as hospital outpatient or physician office visits.

Study 5: Percentage Difference in 30-Day Readmission Rates (Various Denominators), Medicare Advantage vs. FFS

Study 6. Readmission Rates in Medicare --FFS vs. MA

MA vs. FFS Readmission Rates (Jencks Method), All Discharges (Medical and Surgical), Eleven MA Plans (MedAssurant, 2006-2008) and National FFS (Jencks, 2004)

Interval after Discharge

Patients at Risk at Beginning of Period %

Cumulative Rehospitalizations by End of Period

%

Cum. Death or Disenrollment w/o

Rehospitalization by End of Period

%

2004 (FFS/ Jencks)

0-30 days 2,961,460 100.0% 579,903 19.6% 103,741 3.5%

31-60 days 2,277,816 76.9% 834,369 28.2% 134,697 4.5%

61-90 days 1,992,394 67.3% 1,006,762 34.0% 151,901 5.1%

91-180 days 1,802,797 60.9% 1,325,645 44.8% 177,234 6.0%

2006-2008 (11 MA Plans / MedAssurant)

0-30 days 231,085 100.0% 33,641 14.6% 3,393 1.5%

31-60 days 194,051 84.0% 48,931 21.2% 5,288 2.3%

61-90 days 176,657 76.4% 59,488 25.7% 6,560 2.8%

91-180 days 164,789 71.3% 79,890 34.6% 8,951 3.9%

Source: Jencks et. al. (NEJM April 2009), MedAssurant, Inc./AHIP.Notes: Not risk-adjusted.

Study 6. FFS vs. MA (All Enrollees), with Risk Adjustment (DRG Based)Risk-adjusted and Unadjusted Thirty-Day Readmission Rates (Anderson Method),

MA vs. Medicare FFS, 2006-2008

FFS MAPercentage Difference

(MA vs. FFS)

Readmissions per Admission

Unadjusted, All Enrollees* 18.4% 14.4% -22%

Adjustment for DRG Distribution (version 24 DRGs) -- +11.0%

Risk-adjusted 18.4% 16.0% -13%

Adjustment for DRG Distribution (version 26 DRGs) -- +2.7%

Risk-adjusted 18.4% 14.8% -20%

Source: AHIP/MedAssurant based on data from Gerard Anderson/Johns Hopkins University for FFS; MedAssurant, Inc. for MA.*Enrollees under age 65 with end stage renal disease (ESRD) were excluded from the FFS data.

Cost Conference, March 2011

Conference conclusions

1) the need to focus on improved care for patients with chronic illnesses, whose health care needs drive most health spending

2) the potential for sophisticated contracts between health plans and health care systems to improve care for patients with chronic diseases while also slowing the growth of costs

3) the importance of systematically addressing costs throughout the health care system and across the country, not simply focusing on individual payers, providers, or government programs

BCBSMN/FairviewShared incentives for value creation

Year 1 Year 2 Year 3

Incentive payment based on measurable improvements in cost, quality, outcomes

ReduceCost of CareEarned Incentive



Improve QualityEarned Incentive Improve Quality

Earned Incentive Improve QualityEarned Incentive

Fee for ServiceIncreaseGuaranteed



Historical

Fee for ServiceIncrease

Guaranteed

Year 1

Pay for Performance

Chronic Illness– Diabetic care– Vascular care– Hypertension control

Prevention and Wellness– Breast cancer screening– Colorectal cancer

screening– Body Mass Index– Tobacco cessation

Patient Integration– Depression

Safety– Reduction of elective

deliveries– Hospital associated

DVT/PE

Utilization– Preventable events:

Complications, admissions, re-admissions

– Lower back pain– Advanced care

directives

BCBSMN/Fairview Quality Measures

BCBSMNShared incentives for value creation

Year 1 Year 2 Year 3

Incentive payment based on measurable improvements in cost, quality, outcomes




Improve QualityEarned Incentive Improve Quality

Earned Incentive Improve QualityEarned Incentive




Historical

Fee for ServiceIncrease

Guaranteed

Year 1

Pay for Performance

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

Year

ove

r Yea

r % C

hang

e

BCBSMN Medical Cost Trend 8.20% 7.60% 6.30% 6.70% 5.90% 2.50%

Industry Medical Cost Trend 7.40% 7.90% 7.90% 7.10% 7.00% 6.00%

Milliman HCI 9.30% 9.40% 9.00% 8.50% 8.20% 7.00%

Medical Cost Inflation 4.20% 3.60% 5.20% 2.70% 3.40% 3.50%

Blue Trend Survey Median 7.90% 7.60% 6.60% 6.80% 8.20% 6.50%

2005 2006 2007 2008 2009 2010 Est

Fairview Health Services -- Demonstrating Outcomes: Bending the Cost Curve

21

Specific, Clinical Targets?• Preventable and unnecessary hospital readmissions. • Adverse safety events and healthcare acquired

conditions.• Overuse of high-radiation CT scans where ultrasound

or MRI would suffice.• Unmanaged poly-pharmacy in patients with chronic

illnesses.• Overuse of cardiac stents absent heart attack (?)• Overuse of interventions related to prostate cancer

(?)

Medicare FFS 2008 100% Inpatient Dataset

Same Quarter, Same Hospital or Hospital Region, All Cause* Readmissions (No Jencks Possible/Likely Planned Readmission DRGs)

Admits Readmits Rate

Same Hospital Referral Region (HRR) 8,060,588 1,382,827 17.2%

Same Hospital Service Area (HSA) 8,060,588 1,288,462 16.0%

Same Hospital 8,060,588 1,204,847 14.9%

Same-Hospital Readmission Rates in Medicare Fee-for-Service, 2008, Hospitals with the Highest Numbers of Medicare FFS Admissions

HospitalCase Mix Index

CMI wtd avg=1

Total Admits

All-Cause Readmits

Readmit Rate

W/ CMI Adjust

US Total 1.00 8,060,588 1,204,847 14.9%

1 1.61 1.04 32,490 5,799 17.8% 17.1%2 1.89 1.22 22,068 3,269 14.8% 12.1%3 1.72 1.11 20,057 3,346 16.7% 15.0%4 1.61 1.04 19,616 3,060 15.6% 15.0%5 1.65 1.07 19,211 2,690 14.0% 13.1%6 1.82 1.17 17,238 2,903 16.8% 14.3%7 1.60 1.04 17,083 2,600 15.2% 14.7%8 1.61 1.04 16,273 2,595 15.9% 15.3%9 1.55 1.00 16,252 2,096 12.9% 12.9%

10 1.70 1.10 15,755 2,131 13.5% 12.3%Source: Medicare FFS 100% Limited Data Sets; Case Mix Index (CMI) information is from CMS.Note: Case-mix adjusted, all-cause same-quarter readmissions per hospitalization (excluding admissions for rehabilitation and admissions that may be associated with planned readmissions).

Same-Hospital Readmission Rates in Medicare Fee-for-Service, 2008, Hospitals with the Highest Numbers of Medicare FFS Admissions

HospitalCase Mix Index

CMI wtd avg=1

Total Admits

All-Cause Readmits

Readmit Rate

W/ CMI Adjust

US Total 1.00 8,060,588 1,204,847 14.9%

11 1.29 0.84 15,226 2,509 16.5% 19.7%12 1.63 1.06 14,919 2,571 17.2% 16.3%13 1.81 1.17 14,862 2,582 17.4% 14.8%14 1.53 0.99 14,739 3,016 20.5% 20.7%15 1.83 1.19 14,669 2,066 14.1% 11.9%16 1.66 1.07 14,569 2,048 14.1% 13.1%17 1.79 1.16 14,508 1,808 12.5% 10.7%18 1.73 1.12 14,273 2,017 14.1% 12.6%19 1.84 1.19 14,185 2,238 15.8% 13.3%20 2.02 1.31 14,169 1,832 12.9% 9.9%

Source: Medicare FFS 100% Limited Data Sets; Case Mix Index (CMI) information is from CMS.Note: Case-mix adjusted, all-cause same-quarter readmissions per hospitalization (excluding admissions for rehabilitation and admissions that may be associated with planned readmissions).

Same-Hospital Readmission Rates in Medicare Fee-for-Service, 2008, Hospitals with 4,000+ Admissions Ranked by CMI-Adjusted Readmission Rate Hospital

Case Mix Index

CMI wtd avg=1

Total Admits

All-Cause Readmits

Readmit Rate

W/ CMI Adjust

US Total 1.00 8,060,588 1,204,847 14.9%

- 1.65 1.07 5,505 498 9.0% 8.5%- 2.12 1.37 5,237 484 9.2% 6.7%- 1.78 1.15 4,173 422 10.1% 8.8%- 1.97 1.28 7,529 769 10.2% 8.0%- 1.85 1.20 4,830 496 10.3% 8.6%- 1.66 1.07 4,178 430 10.3% 9.6%- 1.78 1.15 4,422 457 10.3% 9.0%- 1.43 0.92 5,178 540 10.4% 11.3%- 1.52 0.98 4,943 527 10.7% 10.8%- 1.59 1.03 4,889 528 10.8% 10.5%


Same-Hospital Readmission Rates in Medicare Fee-for-Service, 2008, Hospitals with 8,000+ Admissions Ranked by CMI-Adjusted Readmission Rate Hospital

Case Mix Index

CMI wtd avg=1

Total Admits

All-Cause Readmits

Readmit Rate

W/ CMI Adjust

US Total 1.00 8,060,588 1,204,847 14.9%

- 1.60 1.03 9,594 1,093 11.4% 11.0%- 1.86 1.21 9,895 1,129 11.4% 9.5%- 2.01 1.30 9,846 1,134 11.5% 8.9%- 2.27 1.47 8,178 946 11.6% 7.9%- 1.90 1.23 8,448 988 11.7% 9.5%- 1.76 1.14 9,250 1,098 11.9% 10.4%- 1.80 1.16 10,374 1,241 12.0% 10.3%- 1.66 1.08 8,445 1,016 12.0% 11.2%- 1.72 1.11 11,644 1,408 12.1% 10.9%- 1.86 1.20 11,712 1,432 12.2% 10.2%


IP_100_2008 -- FFS Only; SGMT_NUM = 1; No Transfers; No Jencks; Select Hospitals -- Same Qtr, Any DRG, Same HRR Readmits

HRR Number HRR City State Admits by HRR ReAdmits by HRRReadmission Rate by HRR

TOTAL US 8,060,588 1,382,827 17.2%422 Provo UT 3,028 286 9.4%370 Rapid City SD 4,173 422 10.1%341 Bend OR 3,090 328 10.6%240 Marquette MI 3,270 349 10.7%421 Ogden UT 3,521 393 11.2%152 Idaho Falls ID 2,706 308 11.4%452 Neenah WI 1,832 209 11.4%002 Dothan AL 6,493 752 11.6%010 Anchorage AK 3,872 451 11.6%315 Hickory NC 6,880 804 11.7%089 Santa Rosa CA 4,638 547 11.8%448 La Crosse WI 5,384 640 11.9%281 Lebanon NH 6,338 754 11.9%276 Missoula MT 5,615 668 11.9%150 Honolulu HI 7,182 868 12.1%345 Salem OR 2,984 362 12.1%105 Grand Junction CO 2,804 343 12.2%073 Redding CA 7,173 880 12.3%440 Spokane WA 26,658 3,277 12.3%151 Boise ID 11,355 1,404 12.4%261 Oxford MS 3,632 453 12.5%016 Fort Smith AR 8,508 1,067 12.5%321 Bismarck ND 5,887 742 12.6%275 Great Falls MT 4,246 536 12.6%423 Salt Lake City UT 19,910 2,515 12.6%

IP_100_2008 -- FFS Only; SGMT_NUM = 1; No Transfers; No Jencks; Select Hospitals -- Same Qtr, Any DRG, Same HRR Readmits

HRR Number HRR City State Admits by HRR ReAdmits by HRR Readmission Rate by HRR TOTAL US 8,060,588 1,382,827 17.2%

223 Baltimore MD 134,849 31,648 23.5%297 Bronx NY 29,811 6,431 21.6%234 Detroit MI 77,772 16,486 21.2%155 Blue Island IL 55,567 11,695 21.0%156 Chicago IL 86,898 18,271 21.0%127 Miami FL 75,203 15,406 20.5%164 Joliet IL 26,304 5,373 20.4%328 Cleveland OH 92,390 18,856 20.4%289 Newark NJ 48,796 9,847 20.2%226 Takoma Park MD 24,754 4,944 20.0%233 Dearborn MI 27,580 5,448 19.8%227 Boston MA 187,626 36,895 19.7%273 St. Louis MO 127,641 24,832 19.5%303 Manhattan NY 140,597 27,232 19.4%357 Pittsburgh PA 84,766 16,407 19.4%356 Philadelphia PA 136,879 26,426 19.3%158 Elgin IL 22,133 4,221 19.1%396 Harlingen TX 14,502 2,764 19.1%364 Providence RI 32,857 6,243 19.0%232 Ann Arbor MI 44,566 8,450 19.0%111 New Haven CT 51,601 9,780 19.0%329 Columbus OH 77,873 14,730 18.9%235 Flint MI 25,426 4,784 18.8%186 Munster IN 19,299 3,631 18.8%331 Elyria OH 10,456 1,966 18.8%

IP_100_2008 -- FFS Only; SGMT_NUM = 1; No Transfers; No Jencks; Select Hospitals -- Same Qtr, Any DRG, Same HSA Readmits

HSA Number HSA City State Total Admits Readmits by HSA Readmission Rate by HSA TOTAL US 8,060,588 1,288,462 16.0%

46025 Provo UT 3,028 286 9.4%03008 Flagstaff AZ 2,823 276 9.8%10033 Englewood FL 2,135 210 9.8%13020 Lewiston ID 1,641 164 10.0%43036 Rapid City SD 4,173 422 10.1%38012 Corvallis OR 1,831 187 10.2%28042 Kearney NE 2,771 284 10.2%38016 Eugene OR 3,714 384 10.3%10016 Cape Coral FL 5,178 540 10.4%10111 South Miami FL 2,747 287 10.4%27034 Missoula MT 2,822 296 10.5%38006 Bend OR 3,090 328 10.6%45096 Fredericksburg TX 1,515 161 10.6%23072 Marquette MI 3,270 349 10.7%52015 Brookfield WI 2,215 237 10.7%01026 Dothan AL 4,889 528 10.8%37080 Shawnee OK 1,989 215 10.8%03015 Lake Havasu City AZ 2,496 271 10.9%50023 Federal Way WA 2,185 238 10.9%38027 Klamath Falls OR 1,922 211 11.0%13038 Twin Falls ID 2,704 297 11.0%10032 Dunedin FL 2,710 298 11.0%05066 Eureka CA 2,304 256 11.1%46020 Ogden UT 3,521 393 11.2%

IP_100_2008 -- FFS Only; SGMT_NUM = 1; No Transfers; No Jencks; Select Hospitals -- Same Qtr, Any DRG, Same HSA Readmits

HSA Number HSA City State Total Admits Readmits by HSA Readmission Rate by HSA TOTAL US 8,060,588 1,288,462 16.0%

25018 Cleveland MS 2,120 495 23.3%37028 Durant OK 3,420 788 23.0%21002 Baltimore MD 93,417 21,469 23.0%36037 East Liverpool OH 2,397 536 22.4%33012 Bronx NY 26,659 5,922 22.2%31038 Orange NJ 3,154 697 22.1%39087 Natrona Heights PA 2,883 632 21.9%44020 Dyersburg TN 3,012 654 21.7%14020 Centralia IL 3,320 717 21.6%05151 National City CA 3,869 835 21.6%31020 Hoboken NJ 2,367 508 21.5%18080 South Williamson KY 1,237 265 21.4%18034 Hazard KY 4,018 860 21.4%14023 Chicago IL 86,898 18,271 21.0%14068 Joliet IL 15,478 3,232 20.9%23026 Detroit MI 27,600 5,692 20.6%31063 Trenton NJ 13,679 2,820 20.6%21022 Leonardtown MD 3,619 741 20.5%33014 Brooklyn NY 48,638 9,952 20.5%39032 Darby PA 4,785 979 20.5%18031 Harlan KY 1,410 288 20.4%05166 Oroville CA 3,534 720 20.4%14024 Chicago Heights IL 9,502 1,928 20.3%11046 Douglas GA 2,098 425 20.3%

Readmission Rates by Hospital Service Area (Medicare FFS 5% Sample Files)

2004 2005 2006 2007 2008 200910%

12%

14%

16%

18%

20%

22%

24%

26%

2004-2009 Readmit Rate per Admit

Los Angeles

Portland

Miami

Baltimore

Houston

Chicago

Boston

Year

Rea

dm

it R

ate

Healthcare Acquired Conditions, Never Events1. Foreign Object Retained After Surgery

2. Air Embolism

3. Blood Incompatibility

4. Stage III and IV Pressure Ulcers

5. Falls and Trauma – Fractures, Dislocations, Intracranial Injuries, Crushing Injuries, Burns, Electric Shock

6. Manifestations of Poor Glycemic Control – Diabetic Ketoacidosis, Nonketotic Hyperosmolar Coma, Hypoglycemic Coma, Secondary Diabetes with Ketoacidosis, Secondary

Diabetes with Hyperosmolarity

7. Catheter-Associated Urinary Tract Infection (UTI)

8. Vascular Catheter-Associated Infection

9. Surgical Site Infection Following: – Coronary Artery Bypass Graft (CABG): Mediastinitis– Bariatric Surgery: Laparoscopic Gastric Bypass, Gastroenterostomy, Laparoscopic Gastric Restrictive Surgery– Orthopedic Procedures: Spine, Neck, Shoulder, Elbow

10. Deep Vein Thrombosis (DVT)/Pulmonary Embolism (PE)– Total Knee Replacement, Hip Replacement

§ Clostridium difficile Infections (CDI)

§ Ventilator-associated Pneumonia (VAP)

§ Methicillin-resistant Staphylococcus Aureus infection (MRSA). V09.0 Infection with microorganisms resistant to penicillins. MRSA is always a secondary diagnosis. The diagnosis co-exists at the time of admission or develops during the stay.

Identifying HACs from Discharge Datasets, NY and CA 2008

All Hospital-Acquired

ConditionsFalls and Trauma

Methicillin-Resistant

Staph aureus (MRSA)

Catheter-Associated

Urinary Tract Infection

Clostridium difficile

Infections

Vascular Catheter-

Associated Infection Other

All Facilities

Total HACs Identified New York 14,171 1,254 1,029 594 8,398 1,692 1,204 California 19,220 1,796 3,429 977 8,023 3,438 1,557

Average per 1,000 Admissions

New York 5.58 0.49 0.40 0.23 3.31 0.67 0.47 California 4.69 0.49 0.93 0.27 2.18 0.94 0.42

Average per 1,000 Days


Hospitals with 8,000 or More Admissions

Lowest Rate per 1,000 Admissions


Highest Rate per 1,000 Admissions


Lowest Rate per 1,000 Inpatient Days


Highest Rate per 1,000 Inpatient Days


Sources: AHRQ NY and CA 2008 SIDs.

Identifying HACs from Discharge Datasets, NY 2008

All Never Events and Hospital-

Acquired Infections Falls and Trauma

Methicillin-Resistant Staph aureus (MRSA)

Catheter-Associated

Urinary Tract Infection

Clostridium difficile infections

Vascular Catheter-

Associated Infection

Total Incidents Identified 14,203 1,254 1,029 594 8,398 1,692Per 1,000 Admissions

Statewide Average 5.59 0.49 0.40 0.23 3.31 0.67Median Among Hospitals in Sample 4.52 0.49 0.32 0.05 2.33 0.27

Lowest Rate per 1,000 Admissions

Overall 0.00 0.00 0.00 0.00 0.00 0.00> 1,000 Admissions 0.00 0.00 0.00 0.00 0.00 0.00> 3,000 Admissions 0.00 0.00 0.00 0.00 0.00 0.00> 5,500 Admissions 0.43 0.00 0.00 0.00 0.10 0.00> 8,000 Admissions 0.43 0.00 0.00 0.00 0.10 0.00> 12,000 Admissions 0.43 0.00 0.00 0.00 0.12 0.00> 16,000 Admissions 0.43 0.00 0.00 0.00 0.12 0.00> 22,000 Admissions 1.18 0.10 0.05 0.00 0.54 0.04

Highest Rate per 1,000 Admissions

Overall 30.09 5.71 7.87 4.44 11.40 25.46> 1,000 Admissions 18.12 2.04 2.31 4.44 11.40 5.25> 3,000 Admissions 18.12 2.04 2.31 4.44 11.40 5.25> 5,500 Admissions 18.12 2.04 2.31 4.44 11.40 5.25> 8,000 Admissions 18.12 1.95 2.31 4.44 11.40 5.25> 12,000 Admissions 18.12 1.41 2.31 4.44 11.40 5.25> 16,000 Admissions 18.12 1.41 2.31 4.44 11.40 5.25> 22,000 Admissions 11.50 1.17 1.53 0.66 6.28 3.77

Source: AHRQ NY 2008 SID. Note: Total events includes rare and “never” events identified in Tables 3 and 4.

Supplemental Q&A Slides

Study 1 -- 8 MA Plans vs. FFS (5%), Compared Individually in the Same Counties

• HMO enrollees only (MA)• 7 Non-profit, one Blue• 2005 and 2006 data (pooled)• 12 month enrollees, age 65-89• Same counties• “CMS-Style” Risk Scores from age, sex, 70 HCCs, serious

diagnoses, primary and secondary diagnoses, inpatient, outpatient, office

• No Medicaid in FFS• FFS 5% sample file coding logics given to plans• 13 “potentially avoidable” admissions logic from AHRQ• Readmissions per enrollee, by DRG in same quarter

Study 1: Eight Company/FFS Sample Basics

Sample Characteristics: HMO Enrollees in Eight MA Plans and FFS Enrollees Nationally and in Local Comparison Areas

Data for 2005 and 2006 (pooled) Average Age (All Enrollees)

Enrollees with Diabetes (HCCs 15-19)

Enrollees with Heart Diseases (HCCs 79-83, 92, 104, 105)

Number of Records Total HCCs Number of

Records Total HCCs

National FFS Including Medicaid 74.2 639,124 2,344,115 859,916 3,342,530 FFS Medicaid 75.1 105,572 468,752 124,536 588,944 FFS Without Medicaid 74.1 533,552 1,875,363 735,380 2,753,586

Company 1 Area MA Plan 74.9 8,031 26,409 8,491 32,322 Local FFS (w/o Medicaid) 73.4 5,916 22,147 8,045 32,300




Study 1: Eight Company/FFS Sample Basics (Continued)

Sample Characteristics: HMO Enrollees in Eight MA Plans and FFS Enrollees Nationally and in Local Comparison Areas

Data for 2005 and 2006 (pooled) Average Age (All Enrollees)

Enrollees with Diabetes (HCCs 15-19)

Enrollees with Heart Diseases (HCCs 79-83, 92, 104, 105)

Number of Records Total HCCs Number of

Records Total HCCs

National FFS Including Medicaid 74.2 639,124 2,344,115 859,916 3,342,530 FFS Medicaid 75.1 105,572 468,752 124,536 588,944 FFS Without Medicaid 74.1 533,552 1,875,363 735,380 2,753,586





Study 1: Eight Company/FFS Severe Diagnosis Rates and Risk Scores

Average Rates of Illness Diagnoses (HCC Groups) and Overall Risk Scores – Medicare FFS and Eight Medicare Advantage HMO Plans

Average HCC Markers per Beneficiary Record, 2005 and 2006 (pooled)

All HCCsCancers

(HCCs 7-10) Diabetes

(HCCs 15-19)

Heart Disease (HCCs 79-

83,92,104-105)

Kidney and Renal

Disease (HCCs 130-

132)

Average Overall

Computed Risk Score *

National FFS Including Medicaid 1.74 0.16 0.34 0.58 0.06 1.13 FFS Medicaid 2.78 0.14 0.59 0.87 0.11 1.61 FFS Without Medicaid 1.62 0.16 0.31 0.54 0.05 1.08

Company 1 Area MA Plan 1.47 0.10 0.36 0.46 0.06 1.03 Local FFS (w/o Medicaid) 1.64 0.14 0.31 0.54 0.06 1.09



Study 1: Eight-Company/FFS Severe Diagnosis Rates and Risk Scores (continued)

Average Rates of Illness Diagnoses (HCC Groups) and Overall Risk Scores – Medicare FFS and Eight Medicare Advantage HMO Plans

Average HCC Markers per Beneficiary Record, 2005 and 2006 (pooled)

All HCCs

Cancers (HCCs 7-10)

Diabetes (HCCs 15-19)

Heart Disease (HCCs 79-

83,92,104-105)

Kidney and Renal

Disease (HCCs 130-

132)

Average Overall

Computed Risk Score *






Sex, Age Risk Factor Value

Female, Age 65-69 0.307

Female, Age 70-74 0.384

Female, Age 75-79 0.483

Female, Age 80-84 0.572

Female, Age 85-89 0.665

Male, Age 65-69 0.346

Male, Age 70-74 0.453

Male, Age 75-79 0.577

Male, Age 80-84 0.657

Male, Age 85-89 0.79

Source: Age/Sex risk factor values are from the 2005 MA Ratebook compiled by the Centers for Medicare and Medicaid Services (CMS).

Age/Sex Risk Incremental Values

HCC Description Risk Factor Value

HCC1 HIV/AIDS 0.685

HCC2 Septicemia/Shock 0.89

HCC5 Opportunistic Infections 0.652

HCC7 Metastatic Cancer and Acute Leukemia 1.464

HCC8 Lung, Upper Digestive Tract, and Other Severe Cancers 1.464

HCC9 Lymphatic, Head and Neck, Brain, and Other Major Cancers 0.69

HCC10 Breast, Prostate, Colorectal and Other Cancers and Tumors 0.233

HCC15 Diabetes with Renal or Peripheral Circulatory Manifestation 0.764

HCC16 Diabetes with Neurologic or Other Specified Manifestation 0.552

HCC17 Diabetes with Acute Complications 0.391

HCC18 Diabetes with Ophthalmologic or Unspecified Manifestation 0.343

HCC19 Diabetes without Complication 0.2

HCC21 Protein-Calorie Malnutrition 0.922

HCC25 End-Stage Liver Disease 0.9

HCC26 Cirrhosis of Liver 0.516

Disease Group Values (15 of 70 HCCs)

INCLUSIONS EXCLUSIONSICD-9-CM Principal Diagnosis Codes

for Bacterial Pneumonia Exclude These Cases

481 (Pneumococcal Pneumonia) Transferring From Another Institution (SID ASOURCE=2) 4822 (H.Influenzae Pneumonia) MDC 14 (Pregnancy, Childbirth, And Puerperium)

48230 (Strep Pneumonia Unspecified) MDC 15 (Newborn And Other Neonates)

48231 (Grp A Strep Pneumonia) With Diagnosis Code For Sickle Cell Anemia Or HB-S Disease (see below)

48232 (Grp B Strep Pneumonia) 48239 (Oth Strep Pneumonia ) Exclude These ICD-9-CM Diagnosis Codes4829 (Bacterial Pneumonia Nos) 28241 (Thalassemia HB-S W/O Crisis)4830 (Mycoplasma Pneumonia) 28242 (Thalassemia HB-S W Crisis)4831 (Chlamydia Pneumonia Oct96-) 28260 (Sickle Cell Disease Nos)4838 (Oth Spec Org Pneumonia) 28261 (HB-S Disease W/O Crisis)485 (Bronchopneumonia Org Nos) 28262 (HB-S Disease W Crisis)486 (Pneumonia, Organism Nos) 28263 (HB-S /HB-C Disease W/O Crisis)

28264(HB-S /HB-C Disease W Crisis)28268 (HB-S Disease W/O Crisis Nec)28269 (HB-Ss Disease Nec W Crisis)

Source: AHRQ Quality Indicators, Prevention Quality Indicators: Technical Specifications; Department of Health and Human Services, Agency for Healthcare Research and Quality, http://www.qualityindicators.ahrq.gov. October 2001. Version 3.1 (March 12, 2007).

Criteria for Identifying Potentially Avoidable Admission for Bacterial Pneumonia

Study 2: California and Nevada (2006) HCUP Sample Basics

Basic Data Characteristics: Records with At Least One Hospital Admission in a Year, Acute Care Admissions

Medicare Advantage Fee-for-Service

Average Number of Records

Average Age

Average Risk

Score*

Average Number of Records

Average Age

Average Risk

Score*

Data for 2006

California – Acute Care Hospitals

All Enrollees 176,794 77.4 1.83 376,721 77.3 1.87

Diabetes Patients 51,749 76.4 2.53 112,739 76.4 2.54

Heart Disease Patients 94,920 78.1 2.40 199,528 78.2 2.49

Nevada – Acute Care Hospitals

All Enrollees 14,396 74.5 1.60 33,862 75.6 1.73

Diabetes Patients 3,894 74.0 2.01 9,161 75.0 2.19

Heart Disease Patients 7,423 75.2 2.17 17,863 76.4 2.31

Source: ARHD SID (H-CUP), Medicaid included in FFS. http://www.ahipresearch.org/pdfs/CAvsNV.pdf

http://www.ahipresearch.org/pdfs/CAvsNV.pdf

Study 4: Alternative Risk Adjustments

Risk Measure (MA vs. FFS) Comparison of Any DRG Readmission Rate, Persons with an Admission (MA vs. FFS)

CMS-Style* Readmission Probability-Based** CMS-Style* Readmission

Probability-Based** California (2006) -2.6% -4.3% -25% -23% California (2007) -2.7% -4.1% -21% -20% Nevada (2006) -8.2% -6.8% -23% -24% Nevada (2007) -10.4% -7.1% -21% -23% Washington (2006) -9.9% -3.5% -20% -26% Texas (2007) -12.8% -4.1% -17% -25% North Carolina (2007) -9.0% -1.6% -16% -22% Pennsylvania (2007) -5.0% -2.2% 0% -3% Hawaii (2007) 1.2% -1.0% -3% -1% Arkansas (2007) -20.3% -6.8% -34% -44% Arizona (2007) -0.1% -0.8% 4% 5%Source: AHIP Center for Policy and Research. Based on state discharge data (HCUP) compiled by the Agency for Healthcare Research and Quality (AHRQ) and the states of Texas and Pennsylvania.Note: The shadings represent our subjective assessment of the reliability of the MA vs. FFS comparisons based on issues with the underlying datasets. In general, we believe the comparisons for California are highly reliable, and the data from Arizona, Arkansas, and Hawaii are not very reliable.* CMS-style risk scores for FFS and MA enrollees based on age/sex and HCC relative cost values used in Medicare risk adjustment for beneficiaries living in the community, but do not include disease interactive factors, or factors related to disability or institutional status. Person-based risk scores are based on inpatient hospital diagnoses for all admissions, but do not include diagnosis information from other health care services, such as hospital outpatient or physician office visits. Relative risk ratios for patients with an admission were used to adjust the per-enrollee results.**Risk measure based on an index of the likelihood of admissions for DRGs that are associated with higher or lower than average rates of readmissions (any DRG).

Study 6. Jencks (FFS) vs. MA

Interval after Discharge Patients at Risk at Beginning of Period %

Cumulative Rehospitalizations by End of

Period%

Cum. Death or Disenrollment w/o Rehospitalization by End

of Period%

2004 (Jencks)0-30 days 2,961,460 100.0% 579,903 19.6% 103,741 3.5%31-60 days 2,277,816 76.9% 834,369 28.2% 134,697 4.5%61-90 days 1,992,394 67.3% 1,006,762 34.0% 151,901 5.1%91-180 days 1,802,797 60.9% 1,325,645 44.8% 177,234 6.0%

2006 (11 MA Plans / MedAssurant)0-30 days 62,012 100.0% 9,069 14.6% 817 1.3%31-60 days 52,126 84.1% 13,288 21.4% 1,323 2.1%61-90 days 47,401 76.4% 16,147 26.0% 1,696 2.7%91-180 days 44,169 71.2% 21,661 34.9% 2,491 4.0%

2007 (11 MA Plans / MedAssurant)0-30 days 75,847 100.0% 11,013 14.5% 1,152 1.5%31-60 days 63,682 84.0% 15,968 21.1% 1,840 2.4%61-90 days 58,039 76.4% 19,496 25.7% 2,374 3.1%91-180 days 53,977 71.2% 26,327 34.7% 3,419 4.5%

2008 (11 MA Plans / MedAssurant)0-30 days 93,226 100.0% 13,559 14.5% 1,424 1.5%31-60 days 78,243 83.9% 19,675 21.1% 2,125 2.3%61-90 days 71,217 76.5% 23,845 25.6% 2,490 2.7%91-180 days 66,643 71.5% 31,902 34.2% 3,041 3.3%Source: Jencks et. al. (NEJM April 2009), MedAssurant, Inc./AHIP.Notes: Not risk-adjusted.

Study 6. Jencks (FFS) vs. MA

Condition at Index DischargeJencks

FFS 30-Day Rehospitalization Rate

MA 30-Day Rehospitalization Rate

Jencks FFS Proportion of All

Rehospitalizations

MA Proportion of All

Rehospitalizations

Medical

All Medical 21.1% 15.8% 77.6% 68.3%

Heart Failure 26.9% 21.1% 7.6% 4.5%

Pneumonia 20.1% 14.1% 6.3% 3.5%

COPD 22.6% 17.6% 4.0% 3.2%

Psychoses 24.6% 15.7% 3.5% 1.2%

GI Problem (Esophagitis) 19.2% 11.9% 3.1% 3.2%

Surgical

All Surgical 15.6% 11.7% 22.4% 31.7%

Cardiac Stent 14.5% 12.5% 1.6% 3.9%

Major Joint Replacement 9.9% 7.6% 1.5% 4.0%

Other Vascular Surgery 23.9% 17.5% 1.4% 1.1%

Major Bowel Procedures 16.6% 14.7% 1.0% 1.3%

Other Hip & Femur 17.9% 14.8% 0.8% 1.2%

Source: Jencks et. al. (NEJM April 2009), MedAssurant, Inc. Notes: Not risk-adjusted.

30 Days, Any DRG (Revisit Data)*

Same Quarter, Any DRG (Original)*

Same Quarter, Any DRG (Revisit Data)

MA Rate vs. FFS Rate (Per Hospitalization, with DRG-Based Risk Adjustment**)

California (2008) -18% -15% -15%

California (2006) -18% -13% -12%

California (2007) -17% -14% -14%

Nevada (2006) -17% -15% -14%

Nevada (2007) -18% -15% -15%

Washington (2006) -17% -17% -17%

Texas (2007) -17% -14% -14%

North Carolina (2007) -12% -15% -14%

Source: AHIP, based on analysis of state hospital discharge public use datasets with encrypted “person” identifiers, provided by AHRQ’s HCUP project (CA, NV, WA, NC,) and by states directly (TX). http://www.ahipresearch.org/pdfs/9State-Readmits.pdf Note: Excludes patients with an admission listing an out-of-state address.*Excludes transfer cases.**Risk measure based on an index of the likelihood of admissions for DRGs that are associated with higher or lower than average rates of readmissions (any DRG), using the 2006-2007 FFS 5 percent sample file as a benchmark (DRG version 24) and the 2008 FFS 5 percent sample file (DRG version 25).

Study 5. Percentage Difference in Risk-Adjusted Readmission Rates (Various Definitions), Medicare Advantage vs. FFS

Studies 4,5,6: DRG-Based Risk Adjustment

Selected DRG(s)

Version 24 DRGs

Share of Admissions

FFS readmit probab-ility

relative to

average (avg=1)

Version 26 MS-DRGs

Share of Admissions

FFS readmit probab-ility

relative to

average (avg=1)FFS MA FFS MA

All Patients

Heart Failure

127 4.85% 5.16% 130%291292293

4.63% 4.56% 126%*

COPD 88 3.12% 3.11% 114%190191192

3.80% 3.57% 100%*

Psychoses 430 3.38% 1.35% 134% 885 3.42% 1.58% 132%Major Hip or Knee Surgery

544 3.53% 3.34% 64% 470 3.27% 3.34% 43%

Other Vascular Surgery

553554

0.94% 0.95% 118%*252253254

0.98% 1.15% 107%*

Major Bowel Surgery

569570

1.01% 1.25% 79%*329330331

1.14% 1.61% 71%*

Study 6, Sample Characteristics

FFS 2006-2008 MA 2006-2008

Age

Patients with a

Hospitalization Percent

Patients with a

Hospitalization Percent

65-69 171,442 21.1 192,714 21.2

70-74 167,831 20.6 198,939 21.9

75-79 170,702 21.1 200,382 22.1

80-84 153,671 18.9 166,985 18.4

85-89 98,414 12.1 100,337 11.1

>89 50,809 6.3 48,347 5.3

Gender Percent

Female 479,911 59.0 502,625 55.4

Male 332,958 41.0 405,079 44.6

Region Percent

West 114,650 14.1 63,146 7.0

Midwest 211,898 26.1 242,984 26.8

South 323,097 39.7 393,045 43.3

Northeast 157,236 19.3 207,992 22.9

Other 5,988 0.7 537 0.1

Total 812,869 100.0 907,704 100.0

Study 6. Version 24 DRG Coding OdditiesTable A-2. Risk Adjustment Calculation Data, DRG Version 24, All Patients

DRG DRG Description

Share of Admissions

FFS readmit probability relative to average (avg

= 1)FFS MA034 OTHER DISORDERS OF NERVOUS SYSTEM W CC 0.26% 0.09% 1.05035 OTHER DISORDERS OF NERVOUS SYSTEM W/O CC 0.07% 0.28% 0.77076 OTHER RESP SYSTEM O.R. PROCEDURES W CC 0.37% 0.11% 1.36077 OTHER RESP SYSTEM O.R. PROCEDURES W/O CC 0.01% 0.35% 1.19089 SIMPLE PNEUMONIA & PLEURISY AGE >17 W CC 3.88% 0.89% 0.92

090 SIMPLE PNEUMONIA & PLEURISY AGE >17 W/O CC 0.29% 2.34% 0.57

092 INTERSTITIAL LUNG DISEASE W CC 0.13% 0.04% 1.16093 INTERSTITIAL LUNG DISEASE W/O CC 0.01% 0.10% 0.38094 PNEUMOTHORAX W CC 0.10% 0.03% 1.07095 PNEUMOTHORAX W/O CC 0.01% 0.11% 0.76096 BRONCHITIS & ASTHMA AGE >17 W CC 0.41% 0.34% 0.80097 BRONCHITIS & ASTHMA AGE >17 W/O CC 0.17% 0.23% 0.54099 RESPIRATORY SIGNS & SYMPTOMS W CC 0.18% 0.27% 0.98100 RESPIRATORY SIGNS & SYMPTOMS W/O CC 0.05% 0.01% 0.67101 OTHER RESPIRATORY SYSTEM DIAGNOSES W CC 0.20% 0.19% 1.04

102 OTHER RESPIRATORY SYSTEM DIAGNOSES W/O CC 0.03% 0.07% 0.55

Study 1: “8 Company”-- MA (HMO) Plan Data and FFS 5% File in Same Local Areas.

Documents

rate patients

diabetes patients

personbased risk scores

heart disease patients

medicare risk adjustment

preliminary ma

risk score value

inpatient hospital diagnoses