1

Readmission Measures

Michele Cournan, DNP, RN, CRRN, FNP, ANP-BC

Speaker Disclosure Statement

Michele Cournan has no industry relationships to disclose.

2

Objectives• Discuss NQF measure #2502, 30-day Post

Discharge All Cause Readmission• Discuss replacement measures – potentially

preventable within stay and unplanned potentially preventable 30-day post discharge readmission

• Explain how the two new measures will be calculated and reported

Acronyms

• PPR – potentially preventable readmission• QRP – quality reporting program• NQF – National Quality Forum• TEP – technical expert panel• MedPAC - The Medicare Payment Advisory

Commission • PAC – post acute care• LTCH – long term acute care hospital• FFS – fee for service

3

Why Do We Need These Measures?

• IMPACT Act – resource use– Potentially preventable 30-day post discharge

readmission

• IRF QRP– Potentially preventable within stay readmission

4

Why Do We Need These Measures?• Hospital readmissions among the Medicare

population are common, costly, and often preventable (Friedman & Basu, 2004, Jencks, Williams, & Coleman, 2009)

• 17%–20% of Medicare beneficiaries discharged from the hospital were readmitted within 30 days (MedPAC, Jencks et al., 2009) – 76% were considered potentially avoidable--

associated with $12 billion in Medicare expenditures

ARN Position

• Serious concerns– Claims data– All cause– No access to patient level data

5

IRF & LTCH 2018 Proposed RulesPropose elimination of NQF 2502

But…..

• Potentially Preventable 30-day Post-discharge Readmission Measure

• Potentially Preventable Within Stay Readmission Measure

• Not NQF endorsed

6

Measure Development

• RTI• TEP

PPR Diagnoses• Adult asthma• COPD• CHF• Diabetes (short term

complication)• Hypertension• Hypotension• Dehydration• Aspiration pneumonia• Influenza• C-diff• Septicemia • GI Hemorrhage

• Fluid/Electrolyte Disorders• Anticoagulant complications• Adverse drug events• Arrhythmia (A-fib/A-flutter)• Intestinal impaction• Pressure ulcers• Chronic Seizures• Bacterial Pneumonia• UTI• Kidney Infection• Cellulitis• Acute Renal Failure• Fractures

7

No TEP Consensus

• Gastrointestinal hemorrhage• Influenza• Infections• Fracture of neck of femur

Clinical Rationale• Some hospital readmissions can be prevented, and that

many of these readmissions occur in the context of PAC, including SNF, IRF, and LTCH.

• For certain diagnoses, proper care and management of patients’ conditions (in the facility or by primary care following discharge) along with appropriate, clearly explained and implemented discharge instructions and referrals, can often prevent a patient’s readmission to the hospital.

• Identifying these PPR conditions will assist healthcare providers’ efforts to improve quality of care and coordination across the care continuum.

8

Clinical Rationale

1) Inadequate management of chronic conditions 2) Inadequate management of infections 3) Inadequate management of other unplanned events 4) Inadequate injury prevention

Potentially Preventable Within Stay Readmission Measure

• Adult Asthma• COPD• Acute Bronchitis• CHF• Diabetes (short term comp)

• Influenza• Hypotension• Hypertension• Bacterial Pneumonia

• UTI• Kidney Infection• C-diff• Septicemia• Dehydration• Electrolyte Imbalance• Acute Kidney Failure• Skin and SQ Infections• Aspiration Pneumonia

9

Potentially Preventable Within Stay Readmission Measure

• Anticoagulant Complications

• Acute Delirium• Intestinal Impaction• Pulmonary Embolism• Fracture

• Arrhythmia (fib/flutter)• Anemia• Pressure Ulcer • Deep Vein Thrombosis• Head Injury

Potentially Preventable 30-day Post Discharge Readmission Measure

• Asthma• Acute Bronchitis• COPD• CHF• Diabetes (short term

comp)• Hypertension• Hypotension• Bacterial pneumonia• UTI• Kidney Infection

• Dehydration• Electrolyte imbalance• Acute kidney failure• Skin/SQ infection• Aspiration pneumonia• Arrhythmia (fib/flutter)• Intestinal impaction• Pressure ulcer• Septicemia • C-diff

10

Key Words

• Potentially Preventable• Unplanned

• Claims based• Admitted

Potentially Preventable

In order for a readmission to be considered potentially preventable, it must be coded as the

principal diagnosis on the readmission claim

11

Unplanned

If a readmission claim contains a code for a procedure that is frequently a planned

procedure, then that readmission is designated to be a planned readmission.

However, the readmission is reclassified as unplanned if the claim also contains a code

indicating one or more acute diagnoses from a specified list

Claims Based

Medicare FFS onlyNo additional data collection

12

Admitted

Excludes urgent care, observation, ED

Calculation• Risk-adjusted PPR rate for each PAC provider

– Derived by first calculating a standardized risk ratio• the predicted number of readmissions at the facility

divided by the expected number of readmissions for the same patients if treated at the average PAC provider.

• The standardized risk ratio is then multiplied by the mean readmission rate in the population (i.e., all Medicare FFS patients included in the measure) to generate the PAC provider-level standardized readmission rate of potentially preventable readmissions.

13

Time Frames

• 30-day post discharge– Excludes day of discharge and day after

• Within stay– Includes day of discharge and day after– Program interruptions and discharges

Denominator

• Risk-adjusted expected number of readmissions for eligible discharges – This estimate includes risk adjustment for patient

characteristics with the facility effect removed– The “expected” number of readmissions is the

predicted number of risk-adjusted readmissions if the same patients were treated at the average PAC provider

14

Data Calculation

• Calculated using two years of data • All IRF stays during the two-year time window,

except those that meet the exclusion criteria, will be included in the measure

• For patients with multiple stays during the two-year time window, each stay will be eligible for inclusion in the measure

• Data from 2012 to 2013 were used for measure development

Risk Adjustment Variables• demographic and eligibility characteristics• principal diagnoses • types of surgery or procedure from the prior

short-term stay• comorbidities• length of stay and ICU/CCU utilization from the

immediately prior short-term stay• number of admissions in the year preceding the

PAC admission • Aggregates of the IRF case-mix groups (CMGs)

15

Denominator Exclusions – Post Discharge

• Death• AMA• Less than 18 years old• SNF/IRF/LTCH to another

SNF/IRF/LTCH or to short term acute care hospital at end of stay

• Prior hospitalization was for non-surgical treatment of cancer

• SNF stays in which the prior hospitalization was for pregnancy

• No short term acute care hospital stay within 30 days prior

• Not continuously enrolled in Part A for 12 months prior and at least 30 days post discharge

• Transferred to a federal hospital

• Care received outside of the US

• SNF/IRF/LTCH stays with data that are problematic

Denominator Exclusions – Within Stay

• AMA• Less than 18 years old• Prior hospitalization was

for non-surgical treatment of cancer

• SNF stays in which the prior hospitalization was for pregnancy

• No short term acute care hospital stay within 1 day prior to IRF admission

• Not continuously enrolled in Part A during stay

• Transferred to a federal hospital

• Care received outside of the US

• SNF/IRF/LTCH stays with data that are problematic

16

Numerator

• Hospital readmissions include readmissions to a short-stay acute-care hospital or an LTCH, with a diagnosis considered to be unplanned and potentially preventable. – Readmissions to inpatient psychiatric facilities are

considered planned and not counted.

Numerator

• Mathematically related to the number of patients in the target population who have the event of a potentially preventable, unplanned readmission during the specific readmission window

17

Huh??• The numerator is the risk-adjusted estimate of

the number of unplanned readmissions. – This estimate starts with the observed

readmissions, and is then risk-adjusted for patient characteristics and a statistical estimate of the PAC provider’s effect, beyond patient case mix.

How will you prepare?

• Consider how many Medicare FFS patients you see

• How can you determine who was readmitted post discharge?– Surveys– HIXNY– Calls

18

PEPPER Reports

HIXNYwww.hixny.org

19

Who will see my data?

Public Reporting

20

IRF Compare Websitehttps://www.medicare.gov/inpatientrehabilitationfacilitycompare/

Questions?