Exploring Mortality Scores: How Mortality Scores Improve Quality Data Pam Hess, MA, RHIA, CDIP, CCS, CPC AHIMA Approved ICD 10 CM/PCS Trainer Vice President, Strategy & Operations This is the Full Title of a Session MedASTUTE Consulting, LLC Phoenix, AZ Susan Schmitz, JD, BSN, RN, CCS, CCDS, CDIP Regional Director Clinical Documentation Improvement Kaiser Permanente Pasadena, CA 1
Mortality Scoring Defined Systems & Dashboards 2
Learning Objectives At the completion of this educational activity, the learner will be able to: Understand mortality scoring and why it is important Discuss expected versus observed mortality calculations Differentiate between various scoring systems Explain the importance of public mortality scorecards Discuss how to develop a program for your CDI Team 3
Mortality Scoring Systems Mortality rate is a measure of the frequency of death occurrence in a defined population during a specified interval of time (CDC, 2018). Mortality scoring system (methodology) refers to a variety of severity scales used to predict hospital mortality using a complex set of criteria for risk adjusted mortality rates. The AHRQ s inpatient quality indictors includes mortality rates for specific procedures and medical conditions. CDC (2018). Epidemiology Glossary. Retrieved from https://www.cdc.gov/reproductivehealth/data_stats/glossary.html AHRQ. (2017). Inpatient Quality Indicators Technical Specifications. Version v7.0 Retrieved from http://www.qualityindicators.ahrq.gov/modules/iqi_techspec_icd10_v70.aspx 4
Example APACHE II Scoring Criteria Rapsang, A. & Shyam, D. (2014). Scoring systems in the intensive care unit: A compendium. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/pmc4033855/ 5
Observed to Expected Calculation Observed mortality: The number of patients who died in the hospital Expected mortality: The expected average of hospital patient deaths if provider case mix is the same as the reference population Risk adjusted rate: The rate the hospital would have if its case mix were the same as the case mix in the reference population O/E mortality ratio: Observed deaths/ expected deaths 0.5 mortality rate is 50% lower than expected 1.0 mortality rate is equal to expected 1.5 mortality rate is 50% higher than expected AHRQ (2018) AHRQ Quality Indicators Toolkit. Retrieved from https://www.ahrq.gov/professionals/systems/hospital/qitoolkit/index.html 6
Inpatient Quality Indicator 12: Coronary Artery Bypass Graft (CABG) Mortality Rate Description: In hospital deaths per 1,000 discharges with coronary artery bypass graft (CABG), ages 40 years and older. Excludes obstetric discharges and transfers to another hospital. Numerator: Number of deaths (DISP=20) among cases meeting the inclusion and exclusion rules. Denominator: Discharges, for patients ages 40 years and older, with any listed ICD 10 PCS procedure code for CABG. Exclusions: Transferring to another short term hospital (DISP=2). MDC 14 (pregnancy, childbirth, and puerperium). With missing discharge disposition (DISP=missing), gender (SEX=missing), age (AGE=missing), quarter (DQTR=missing), year (YEAR=missing) or principal diagnosis (DX1=missing). AHRQ. (2017). Inpatient Quality Indicators Technical Specifications. Version v7.0 Retrieved from http://www.qualityindicators.ahrq.gov/modules/iqi_techspec_icd10_v70.aspx 7
Importance of Mortality Scoring Medical decision making based on objective criteria and not subjective judgements Refined scores for comparison purposes with various groups of patients Clinically relevant outcomes and high predictive power Decreased hospital costs Statistical formulas to compare hospitals against peers Severity scores transformed into a hospital death probability using logistic regression equations Rapsang, A. & Shyam, D. (2014). Scoring systems in the intensive care unit: A compendium. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/pmc4033855/ 8
Mortality Rates in Quality Measurement Challenges to accurate quality measurement Clinical record does not reflect severity of illness; secondary diagnoses may be other than MCC/CC. Diagnoses codes impacting severity of illness scores are not intuitive; there is no list. Risk factors may be under or over reported causing biased reporting. Statistical model bias without adjustment for over or under risk factor estimation. Solutions Hospitals should carefully evaluate coding consistency, completeness and accuracy, including the coding of UB 04 fields like admission source and admission status, as well as ICD 10 CM diagnosis and procedure coding. Meurer, S. (2008). Mortality Risk Adjustment Methodology for University Health System s Clinical Data Base University Health System Consortium. Retrieved from https://archive.ahrq.gov/professionals/quality patient safety/quality resources/tools/mortality/meurer.pdf. 9
Mortality Scoring Methodologies 1. HSMR2 (Hospital Standardized Mortality Ratio v2) 2. ACA (Acute Care Admission Mortality Ratio) 3. UHC (University Hospital Consortium) 4. APACHE (Acute Physiology, Age and Chronic Health Evaluation Systems) 5. SAPS (Simplified Acute Physiology Score) 6. MPM (Mortality Prediction Models) 7. POSSUM (Physiological and Operative Severity Score for the enumeration of Mortality and Morbidity) 10
11
Polling Question 1 Which of the following mortality scoring methodologies does your facility use? HSMR2 (Hospital Standardized Mortality Ratio v2) UHC (University Hospital Consortium) APACHE (Acute Physiology, Age and Chronic Health Evaluation Systems) Other I don t know 12
Mortality Scoring Methodology: Drill Down HSMR@ Logistic regression model to calculate the risk of death for each inpatient stay Adjusted for age band, sex, admission type, comorbidity, transfer in, diagnosis subgroup, admission quarter, and an interaction between age band and comorbidity Risks summed together to derive the total expected number of deaths at a hospital in a specified period of time HSMR2 = Observed number of inpatient deaths/expected number after adjustment Restricted to a maximum of 33 CCS diagnosis groups that together contribute to 80% of in hospital deaths in all payer, all ages of the US Healthcare Cost and Utilization Project (HCUP) data in 2012 13
Mortality Scoring Methodology: Drill Down UHC Assignment of a Severity of Illness (SOI) and Risk of Mortality (ROM) level to each case. SOI is defined as the extent of organ system loss of function (3M, 2016). ROM is the likelihood of dying in the hospital (3M, 2016). The ROM class is used for the evaluation of patient mortality. Selection of a patient population to serve as the basis of the model to provide norms. Use of regression techniques to predict probability of mortality based on the normative patient population. Assignment of expected mortality probability to every patient in UHC s clinical database. 3M. (2016). 3M All Patient Refined Diagnosis Related Groups (APR DRGs). Retrieved from https://www.forwardhealth.wi.gov/kw/pdf/handouts/3m_apr_drg_presentation.pdf Meurer, S. (2008). Mortality Risk Adjustment Methodology for University Health System s Clinical Data Base University Health System Consortium. Retrieved from https://archive.ahrq.gov/professionals/quality patient safety/quality resources/tools/mortality/meurer.pdf 14
UHC Independent Variables for Regression Model Independent Variables ROM subclass for mortality, and SOI subclass for LOS and costs Patient age Patient sex Admit source = transfer from another acute care hospital Admit source = transfer from skilled nursing facility Admit source = long term care facility Low SES (based on Medicaid, self pay, charity as primary payer) Admit status = emergency Patient Race Example Comorbid conditions defined by the AHRQ Valvular disease Pulmonary circulation disorders Peripheral vascular disorders Hypertension (complicated and uncomplicated) Paralysis Other neurological disorders Chronic pulmonary disease Diabetes (complicated and uncomplicated) Hypothyroidism Renal failure Liver disease Peptic ulcer disease excluding bleeding Meurer, S. (2008). Mortality Risk Adjustment Methodology for University Health System s Clinical Data Base University Health System Consortium. Retrieved from https://archive.ahrq.gov/professionals/quality patient safety/quality resources/tools/mortality/meurer.pdf 15
Public Scorecards Why Are They Important? Available to the public via the Internet Reflects quality of patient care and is used by perspective patients to shop for hospitals and physicians Used by CMS, The Joint Commission, Healthgrades, and Leapfrog Group to measure quality of care Used by CMS for Value Based Purchasing benchmarks: Example Clinical Care Domain VBP Mortality Measures MORT 30 AMI Acute Myocardial Infarction (AMI) 30 Day Mortality Rate MORT 30 HF Heart Failure (HF) 30 Day Mortality Rate MORT 30 PN Pneumonia (PN) 30 Day Mortality Rate CMS. (2012). Hospital Value Based Purchasing Program. Retrieved from https://www.cms.gov/medicare/quality Initiatives Patient Assessment Instruments/Value Based Programs/HVBP/HVBP FAQs.pdf. 16
AHRQ Quality Indicator Toolkit Dashboard AHRQ (2016) AHRQ Quality Indicators Toolkit. Retrieved from https://www.ahrq.gov/professionals/systems/hospital/qitoolkit/index.html 17
CMS Hospital Compare CMS. (2018) Hospital Compare Retrieved from https://www.medicare.gov/hospitalcompare/search.html? 18
CMS Hospital Compare COPD In the public eye: The death rates show whether patients died within 30 days of surgery after being hospitalized for a specific condition. Graph shows whether a hospital's death rate is better than, no different than, or worse than the national rate. Death rates provide information about important aspects of hospital care that affect patients outcomes like prevention of and response to complications, emphasis on patient safety, and the timeliness of care. CMS. (2018) Hospital Compare Retrieved from https://www.medicare.gov/hospitalcompare/search.html? 19
CMS Variation in 30 Day Risk Standardized Mortality Rates (RSMRs) CMS (2018) Outcome Measures. Retrieved from https://www.cms.gov/medicare/quality Initiatives Patient Assessment Instruments/HospitalQualityInits/OutcomeMeasures.html 20
Initiating a CDI Mortality Program A Team Approach 21
Polling Question 2 How many of you have a CDI Mortality Program in place? Yes No Sort of I don t know 22
Where to Start Obtain approval Executive leader Steering committee Direct manager Obtain buy in from those involved HIM management Billing Develop a mortality report 23
Mortality Report Patient name Patient MRN Birthdate Age at discharge Gender Hospital account number Coding status Coding date Medical center name Attending provider name Attending provider type Admission date Discharge date SOI ROM 24
Designing Your Program Factors to consider Average mortalities per day with SOI/ROM < 4 Resources available Experienced CDI specialist Size of health system Coding and billing cycles 25
Pilot Program What medical center had the most opportunity? Focused review Easily managed Quickly added additional medical centers Within 6 months all medical centers being reviewed 26
Spreadsheet Example 27
Case Study PDX Opportunity Initial SOI/ROM 3/3 Patient s initial PDX was Acute Respiratory Failure however patient was also admitted with marked leukocytosis with WBC > 200. Bone marrow biopsy confirmed Chronic Myelogenous Leukemia (CML). Request was made to coding supervisor to resequence PDX to CML. MS DRG 840, Lymphoma and non acute leukemia with MCC was coded and billed. Subsequent SOI/ROM 3/4 28
Case Study Query Opportunity Initial SOI/ROM 4/2 Patient admitted with Chron s disease and initial MS DRG was 386, Inflammatory Bowel Disease with cc. Noted in ED notes recent pulmonary embolism with current heparin administration. Physician queried and agreed patient had acute PE. Added to coding, PDX changed to MS DRG 385, Inflammatory Bowel Disease with MCC. Subsequent SOI/ROM 4/3 29
Case Study Query Opportunity Initial SOI/ROM 1/1 Patient admitted through ED after fall. Had nausea and started vomiting. Became hypotensive. Systolic pressure in the 60 s. Also c/o chest pain. EKG abnormal. CT head normal. Documentation of hypoxemia, O2 sat s in the 80 s. Became bradycardic and nonresponsive in ICU. Went into asystole. Queries for Acute respiratory failure and acute MI. Physician agreed and placed addendum in the discharge note. Subsequent SOI/ROM 4/4 30
Case Study Coding Opportunity Initial SOI/ROM 3/3 Patient admitted with stroke. MS DRG 065, Intracranial hemorrhage with CC. Glasgow Coma scale documented separately on neuro nurses flow sheet. Coma can be added to coding. Final MS DRG 064, Intracranial hemorrhage with MCC. Subsequent SOI/ROM 3/4 31
2018 Official Guidelines for Coding and Reporting FY 2018 (October 1, 2017 September 30, 2018) 32
Individual Coma Score For Traumatic Brain Injury, if it is the PDX individual GCS scores are not an MCC For CVA, individual GCS scores are MCC If only the total GCS score is available, query for coma (R4020) Total GCS score when coded does not have an impact if CVA is the PDX Unconsciousness codes to unspecified Coma (R4020) If the documentation reflects Persistent Vegetative State (CC) and the GCS score falls between 3 8, need to clarify for Coma (MCC) 33
Results Example 34
Appendix Workflow Design Example 35
Design Example 36
Thank you. Questions? pam@medastute.com Susan.B.Schmitz@kp.org In order to receive your continuing education certificate(s) for this program, you must complete the online evaluation. The link can be found in the continuing education section at the front of the program guide. 37