Upload
hoangdien
View
219
Download
5
Embed Size (px)
Citation preview
Objectives
• Understand the meaning of HCC • Understand how coding and clinical documentation
improvement affect HCCs • Understand the financial impact of HCCs • Learn the history of HCCs and how they are being used
today in population health management and value based purchasing
• Understand the connection between risk stratification of patient populations and potentially competing documentation and coding guidelines for MS-DRGs, APR-DRGs, SOI/ROM, PSIs, HACs and HCCs
Hierarchical Condition Categories (HCCs)
There are two types of HCC’s CMS-HCCs (Medicare) • Developed by CMS for risk
adjustment of the Medicare Advantage Program (Medicare Part C) to predict non-drug medical spending
• CMS also developed a CMS RX HCC model for risk adjustment of Medicare Part D population
• Based on aged population (over 65)
HHS-HCCs (Commercial) • Developed by the Department of
Health and Human Services (HHS) • Designed for the commercial payer
population • HHS-HCCs predict the sum of medical
and drug spending • Includes all ages
CMS-HCCs • Developed by CMS to adjust Medicare
capitation payments to Medicare Advantage Plans (Medicare Part C) based on the health expenditure risk of their enrollees
• Current year data predictive of future year risk
• Based on diagnoses patient on Medicare has accumulated over a year from data submitted from:
– Principal diagnosis hospital inpatient – Secondary diagnosis hospital inpatient – Hospital outpatient diagnoses – Physician office diagnoses – Clinically-trained non-physician
(psychologist or podiatrist) • CMS-HCC will be used in Value Based
Purchasing to determine part of the risk adjustment score
CMS HCC risk adjustment model is prospective with a base year of demographic information combined with major medical conditions to predict Medicare expenditures/risk in the next year.
Total Performance Score for Value Based Purchasing FY 2015
• Clinical process includes: – 12 Process of care measures related
to AMI, HF, PN, SCIP
• Patient experience includes: – Survey of patient experience of care
• Outcome includes: – 30 day Mortality for AMI, HF, PN – PSI 90 – composite of 8 Patient Safety
Indicators – CLABSI – Central Line Associated
Blood Stream Infection
• Efficiency includes: – Medicare Spending per Beneficiary
measure (Includes risk adjustment variables such as HCCs)
AMI = Myocardial Infarction PN = Pneumonia HF = Heart Failure SCIP = Surgical Care Improvement
HHS-HCCs Commercial HCC
• The Department of Health and Human Services (HHS) have also created HCCs to adjust risk in the individual and small group markets.
• Models have been developed by: – Age group (adult, child and infant) – At each cost sharing level (platinum,
gold, silver and bronze metal levels as well as catastrophic plans)
• CMS-HCCs were used as the starting point for these particular categories and adapted for Affordable Care Act (ACA) risk adjustment
• Reasons for the differences between CMS-HCC and HHS-HCC include:
– Population is based on all ages – not just the aged
– Current year diagnoses and demographics used to predict this year risk
– Prediction includes the sum of medical and drug spending
• Map ICD diagnoses to HHS condition categories (CCs) • Subset of ICD codes are used, all others ignored and duplicate diagnosis codes are ignored • Rules for allowed sources apply (hospital, outpatient facility, or professional) - based on
service type - designated by allowed CPT/HCPCS codes - or bill type
• For a given claim/encounter record either all the diagnoses or none will be allowed • Bundled mother/infant claims need to be unbundled
• Map to hierarchical condition categories (HCCs) • Most severe CC assigned from same category
• Compute risk score from the risk adjustment models: • Age/Sex and Metal level
• Risk factors added to produce risk scores
Calculating Risk Scores – HHS-HCCs
Demographics
Diagnoses
Risk Score
CSR Adjusted Risk Score
Source: Demographics • Enroll ID; Age first, Age last; Sex; • Metal level (platinum, gold, silver, bronze, catastrophic) • Cost-Sharing Indicator
Source: Diagnoses • Diagnoses from claims or encounter records for current benefit
year • Based on discharge date or service date
Members without claims or encounter records are assigned zeroes
CMS-HCCs • A risk score of 1.0 reflects the Medicare-
incurred expenditures of an average beneficiary
• Risk adjustment incorporates diagnostic and demographic data
• Demographic data includes, for example, age/sex group, Medicaid status, disability status and if living in an institution
• 79 Categories identified in 2014 – ICD-9 based
• Within each category, there are hierarchies that represent more advanced and costly conditions in a higher coefficient
• There is a formula to account for disease interaction and disabled status
• HCC Category examples: – Infection – Neoplasm – Diabetes – Metabolic – Liver – Gastrointestinal – Musculoskeletal – Blood – Substance Abuse – Psychiatric
Diabetes Category – Need To Get To the Highest Level
HCC 17 Risk 0.474
Institutional
• Diabetes without Complication
HCC 18 Risk 0.474
Institutional
• Diabetes with Chronic Complications
HCC 19 Risk 0.182
Institutional
• Diabetes with Acute Complication
HCC Classification System
ICD-9-CM Codes-14,000+ codes ICD-10-CM Codes – 70,000+ codes
Diagnostic Groups- 805 groups
Condition Categories- 189 categories
Hierarchical Condition Categories (HCC’s)- 189 categories
CMS-HCC- 79 categories used in payment model as of 2014
Goal For Each Patient • Report all current diagnoses at the
highest level of specificity based on physician documentation
• The more categories of diagnoses reported over a year creates a higher risk score
• Only one diagnosis per category is reported
– Angina is not reported as AMI is more severe condition in the category
• While all ICD-9 diagnosis codes fall within an HCC category, only 79 categories are included in the payment model
Clinical vignette for CMS-HCC classification, community-residing, 76-year-old woman with AMI, angina pectoris, COPD, renal failure, chest pain, and ankle sprain
Diagnoses Must Be Documented
• “In order to be acceptable for risk adjusted payments, the risk assessment must be conducted as a face-to-face encounter by a provider that is an acceptable risk adjustment provider type (e.g., a physician or nurse practitioner), and be documented in a medical record in order to meet risk adjustment rules. Examples of diagnoses that could be identified include chronic conditions, such as diabetes and vascular disease.”
• Must meet the UHDDS definitions for reporting diagnoses including: – Evaluation – Treatment – Monitoring
Financial Impact Example • Hypothetical example of CMS-HCC
expenditure predictions and risk score for a community-residing, 76 year old woman with AMI, angina pectoris, COPD, renal failure, chest pain and ankle sprain
• If this same patient had a diabetic chronic complication, that risk factor score would be added to the total
Risk Adjustment Factor (RAF)
Total score of all relative factors related to one patient for a total year • Age and whether community based on institution based • Medicaid Disability and interaction with age and sex • HCC category, based on diagnoses reported • Interaction between certain disease categories • Interaction between certain disease categories and disability status
Note: In 2014 CMS added a normalization factor to account for coding intensity
Providers Are Looking For Ways To Accurately Report HCCs
• Review of this case indicates a patient with both diabetes and Chronic Kidney Disease Stage IV
• Physician must establish the cause and effect relationship between the two conditions in order for a combination code to be reported
• Query for relationship – if answer links the CKD to the Diabetes, the HCC score changes
Same Patient as In Last Slide But Institution Based
• An institution based patient has different relative factor scores • In addition there are disease interaction scores that are added in some cases on both
community based and institution based patients • There is a disease interaction between epilepsy and schizophrenia in an institution which
adds 0.452
Reporting Issues
Diagnoses not always reported
• Secondary Cancers • Diabetic manifestations • Morbid Obesity/BMI >40 • Drug dependence • Hemiplegia due to stroke • Monoplegia or paralysis due
to stroke • Status amputations • Status ostomy
Issues affecting accurate reporting • The majority of patients are only
seen in a physician office setting • Physician office notes are limited
and specificity is not always identified
• Physician’s have to report what is treated, evaluated and monitored and might not be aware of what falls into each category
• Physician’s report on a 1500 claim form which has space only for 4 diagnoses
• Unspecified and symptom diagnoses are not considered as HCCs
Focus Has Been on Inpatient
• Historically providers have focused documentation and reporting of diagnoses based on inpatient care
• With the advent of other reportable conditions that also impact reimbursement, such as HCC’s, HAC’s, PSI’s, etc., providers are looking for new methods to capture data
• Consider a more focused CDI review and coding programs geared to the OP setting
Across the Continuum of Care
• Typically there are not trained coders in physician offices or in some hospital clinics
• As these programs expand across the continuum of care, hospitals will need to focus attention on the accuracy of data reported from inpatient, outpatient, clinics and physician offices.
• CMS has hinted for years that this data will be merged and compared – Prepare!
Goal of Reporting Diagnoses
• Accurate and complete reporting of all conditions
• Tell a complete patient story
As population health and value based purchasing take hold, there will be more interest in accurate reporting.
References
• http://www.cms.gov/Medicare/Health-Plans/MedicareAdvtgSpecRateStats/Risk-Adjustors.html
• https://www.cms.gov/.../HospVBP_FY15_NPC_Final_03052013_508.pdf