Do Hospital Penalties Affect Health Disparities? The Impact of HRRP - - PowerPoint PPT Presentation

June 2017

Do Hospital Penalties Affect Health Disparities?

The Impact of HRRP on Black Patients and the Hospitals that Serve Them

Jose F. Figueroa, MD, MPH

Associate Physician, Brigham & Women’s Hospital Instructor of Medicine, Harvard Medical School @joefigs2

No personal conflict of interests to disclose

Co-authors: Ashish K. Jha, MD, MPH Arnold Epstein, MD, MPH, Jie Zheng, PhD, John Orav, PhD

Background: Hospital Readmissions

 Hospital Readmissions are common, costly, and potentially

preventable

 Over 1 in 5 Medicare pts readmitted (pre-ACA)  ~24% to 76% potentially preventable  Cost: $17 to $24 billion

 Black patients have a 20% higher rate of readmissions than Whites  Care for blacks is highly concentrated (“minority-serving” hospitals)

 Top 10% MSHs: ~ 50% of blacks  Top 25% MSHs: ~ 90% of blacks

 Historically, MSHs have higher readmission rates for all patients

Joynt et al., JAMA, 2011; MedPAC Report, 2007; Jha et al, Health Affairs 2011, Jha et al, JAMA IM, 2007

Hospital Readmissions Reduction Program

Medicare Hospital Penalty Program

• Penalty for higher-than-expected readmission rates

Targeted Conditions

• FY 2013: AMI, heart failure, pneumonia
• FY 2015: COPD, THR/TKR
• FY 2017: CABG

Penalty Size

• FY 2013: 1%
• FY 2015 to present: 3%

Readmission rates are decreasing over time

Zuckerman et al, NEJM, 2016

Can hospital penalties worsen health disparities?

MSHs report more barriers yet less likely to use readmission reduction strategies MSHs are much more likely to be penalized Increase in financial penalties may limit MSHs ability to invest in quality improvement efforts Important to understand impact of hospital penalties on minority populations and hospitals that serve them

Joynt KJ, et al., JAMA, 2011; Joynt KJ, et al., AJMC, 2016, Figueroa JF, et al., Med Care 2017.

Research questions

 Q1. How did the HRRP affect readmission rates of black patients relative to white patients? Q2. Did the HRRP affect the disparity gap between MSHs

• vs. non-MSHs?

Q3. If so, did the risk of MSHs receiving a hospital penalty change over time?

Methods  Data:

• 2007 to 2014 Medicare 100% inpatient file
• Hospital characteristics: AHA survey
• Penalty data: CMS Hospital Compare

Exclusions:

• Hospitals not in 8yrs of period
• Critical access hospitals & MD Hospitals
• Specialty hospitals (children’s, VA hospitals)

Definition: “Minority-serving” hospital

• Top 10% of hospitals with the highest proportion of black

Medicare admissions

Statistical Methods

Analysis: Linear spline regression model

• Time 1: Pre-ACA/HRRP

Q1 2007 to Q1 2010

• Time 2: HRRP implementation

Q2 2010 to Q2 2012

• Time 3: Penalty phase

Q3 2012 to Q4 2014

Outcome: 30-day readmission rates (composite AMI, CHF, PN) Primary Predictors:

• Q1: Race (White vs. Black)
• Q2: Hospital status (MSH vs. non-MSH)

Covariates:

• Age, sex, dual status, comorbidities
• Hospital fixed effects

Results

Patient Characteristics (2007) Blacks

(n=83,003)

Whites

(765,416) Age (mean)

77.1 79.6

Gender

Female 59.8% 53.5%

Comorbidities

Hypertension 63.2% 53.2% Diabetes 31.3% 22.4% COPD 33.2% 39.7% Renal Failure 35.7% 22.1% Obesity 4.7% 2.9%

Patient Characteristics of Blacks vs. Whites

Note: Pattern of patient characteristics was similar across all years in study period (2007 to 2014)

Hospital Characteristics MSH (n=290) Non-MSH (n=2,650)

Hospital Size

Small 22.5% 29.2% Medium 52.6% 56.5% Large 24.9% 14.3%

Teaching Status

Major Teaching 21.5% 6.4% Minor Teaching 29.4% 27.9% Non-Teaching 49.1% 65.7%

Hospital Region

Northeast 13.8% 16.3% Midwest 19.4% 23.5% South 61.2% 39.9% West 5.5% 20.3%

Hospital Characteristics

0% 5% 10% 15% 20% 25% 30%

2007 2008 2009 2010 2011 2012 2013 2014

White Black

Trends in 30d readmission rates in Blacks v Whites

30d composite readmission rate

(AMI, CHF, PN)

Time 1: Pre-ACA Time 2: Implementation Time 3: Penalty Phase

Blacks: slope, 0.06% per quarter slope, -0.45% slope, -0.10% Whites: slope, 0.02% per quarter slope, -0.36% slope, -0.11% Difference-in-differences in Trends

slope, -0.13% per quarter (p<0.001)

(more reduction in Blacks)

Difference-in-differences in Trends

slope, 0.08% per quarter (p=0.10)

24.5% 22.3% 22.5% 23.9% 20.3% 19.3% 19.3% 18.4%

0% 5% 10% 15% 20% 25% 30%

2007 2008 2009 2010 2011 2012 2013 2014

MSH non-MSH

30d composite readmission rate (AMI, CHF, PN)

Time 1: Pre-ACA Time 2: Implementation Time 3: Penalty Phase

Trends in readmission rates in MSHs vs. non-MSHs

Difference-in-differences in Trends

slope, -0.10% per quarter (p=0.009)

(more reduction in MSHs)

Difference-in-differences in Trends

slope, 0.06% per quarter (p=0.18)

MSHs: slope, 0.04% per quarter slope, -0.44% slope, -0.10% Non-MSHs: slope, 0.02% per quarter slope, -0.36% slope, -0.12%

Did the risk of receiving a readmissions penalty from HRRP change over time for MSHs?

Fiscal Year % MSHs Penalized

(n=290)

Odds Ratio of MSH Receiving Penalty*

(95% CI) [non-MSH ref group]

P-value 2013 84.8% 1.5

(1.3 to 1.8)

<0.001

2014 87.3% 1.7

(1.4 to 2.0)

<0.001

2015 90.5% 1.3

(1.1 to 1.6)

0.016

2016 90.5% 1.3

(1.1 to 1.6)

0.016

2017 91.2% 1.4

(1.2 to 1.7)

<0.001

*Multivariate logistic regression model, adjusting for hospital characteristics

Limitations

 Observational study so cannot draw causal inferences between HRRP and changes in readmission rates Administrative claims data are limited (lack of clinical data)

Discussion/Conclusion

 The readmission gap between blacks and whites narrowed after the announcement of HRRP Penalties did not lead to further reductions in disparity gap Similarly, MSHs narrowed readmission gap relative to non-MSHs during HRRP implementation period only Concerns remain since MSHs continue to be more likely penalized than non-MSHs despite experiencing more improvement

Policy Implications

Policymakers should consider changes to HRRP program to reward improvement and not just achievement 21st century act: Congress mandated HHS to account for social risk factors

Adjustment for dual status

THANK YOU

QUESTIONS? JOSE F. FIGUEROA, MD, MPH jfigueroa@hsph.harvard.edu @joefigs2 ACKNOWLEDGMENTS

• Ashish K. Jha, MD, MPH
• Arnold Epstein, MD, MPH
• John Orav, PhD
• Jie Zheng, PhD
• 42 Church St Team at HGHI

Extra Slides/Notes

Formulas to Calculate the Readmission Adjustment Factor

Excess readmission ratio = risk-adjusted predicted readmissions/risk-adjusted expected readmissions Aggregate payments for excess readmissions = [sum of base operating DRG payments for AMI x (excess readmission ratio for AMI-1)] + [sum of base operating DRG payments for HF x (excess readmission ratio for HF-1)] + [sum of base operating DRG payments for PN x (excess readmission ratio for PN-1)] + [sum of base operating DRG payments for COPD x (excess readmission ratio for COPD-1)] + [sum of base operating payments for THA/TKA x (excess readmission ratio for THA/TKA -1)] *Note, if a hospital’s excess readmission ratio for a condition is less than/equal to 1, then there are no aggregate payments for excess readmissions for that condition included in this calculation. Aggregate payments for all discharges = sum of base operating DRG payments for all discharges Ratio = 1 - (Aggregate payments for excess readmissions/ Aggregate payments for all discharges) Readmissions Adjustment Factor = the higher of the Ratio or 0.97 (3% reduction). (For FY 2013, the higher of the Ratio or 0.99% (1% reduction), and for FY 2014, the higher of the Ratio or 0.98% (2% reduction).)

Formulas to Compute the Readmission Payment Adjustment Amount

Wage-adjusted DRG operating amount* = DRG weight x [(labor share x wage index) + (non-labor share x cola, if applicable)]

*Note, If the case is subject to the transfer policy, then this amount includes an applicable

payment adjustment for transfers under § 412.4(f). Base Operating DRG Payment Amount = Wage-adjusted DRG operating amount + new technology payment, if applicable. Readmissions Payment Adjustment Amount = [Base operating DRG payment amount x readmissions adjustment factor] - base operating DRG payment amount.

*The readmissions adjustment factor is always less than 1.0000, therefore, the

readmissions payment adjustment amount will always be a negative amount (i.e., a payment reduction).