Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Specialty > Risk Adjustment > Flashcards

Risk Adjustment Flashcards

1
Q

Factors when building a clinical algorithm

A
  1. Diagnoses
  2. Source of the diagnosis (claims, lab values, medical charts, etc.)
  3. If source is claims, what claims should be considered?
  4. If claim contains more than one diagnosis, how many diagnoses will be considered for identification?
  5. Over what time span will a diagnosis have to appear to be incorporated in the algortihm?
  6. What procedures may help determine the level of severity of diagnosis?
  7. What prescription drugs may be used to identify conditions?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Challenges when constructing a condition-based model

A
  1. Impact of co-morbidities for conditions often found together
  2. Type of benefit design underlying the data
  3. Degree of certainty with which the diagnosis has been identified
  4. Extent of “coverage” of the data (e.g. self-reported data may not be complete)
  5. Level of severity at which to recognize the condition
  6. Large number and different types of codes for procedures and drugs

HINT: CB DESC (Condition Based Description)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Definitions of sensitivity and specificity

A

When building clinical identification algorithms, the proper balance between sensitivity and specificity must be found.

  1. Sensitivity - the percentage of members correctly identified as having a condition (“true positives”)
  2. Specificity - the percentage of members correctly identified as not having a condition (“true negatives”)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

External sources of clinical identification algorithms

A
  1. NCQA/HEDIS
  2. Population Health Alliance
  3. CMS
  4. Quality reporting and improvement organizations
  5. Grouper models
  6. Literature
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Uses of HEDIS data

A
  1. Select the best health plan for their needs (by employers, consultants, and consumers)
  2. Health plans seeking accreditation
  3. Plans participating in Medicare must submit data on HEDIS-developed measures of quality
  4. Many state governments require plans participating in Medicaid to report HEDIS data
  5. May be used in “Pay for Performance” programs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Major publishers of health care quality measures (used for sources of algorithms)

A
  1. National Quality Forum (NQF)
  2. Agency for Healthcare Research and Quality (AHRQ)
  3. Joint Commission
  4. CMS
  5. Hospital Quality Alliance (HQA)
  6. Measures Applications Partnership (MAP)
  7. American Medical Association Physician Consortium for Performance Improvement (PCPI)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Reasons for using commercially-available grouper models

A
  1. Building algorithms from scratch requires considerable time and work
  2. Models must be maintained to accommodate new codes
  3. Risk adjustment, providers, and plans may require a commercially available, consistent model to be used, and it must be available for review and validation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Principles of Grouper Model Design

A
  1. Diagnostic categories should be clinically meaningful
  2. Diagnostic categories should predict medical expenses
  3. Diagnostic categories should have adequate sample sizes for stable estimates
  4. Hierarchies should be used to characterize illness level within each disease process
  5. Diagnostic classification shyould encourage specific coding
  6. Diagnostic classification should not reward coding proliferation
  7. Providers should not be penalized for recording additional diagnoses
  8. Classification system should be internally consistent
  9. Diagnostic system should assign all codes (ICD-9/10)
  10. Discretionary diagnostic categories should be excluded
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Commercially-available diagnosis-based grouper models

A
  1. DRG - case mix to hospital costs incurred
  2. CMS-HCC - condition categories
  3. HHS-HCC - ACA normalization
  4. CRG - Clinical Risk Groups - identifies total resources
  5. CDPS Chronic Illness and Disability Payment System - Adjusting payments (Medicaid, disabled, TANF)
  6. CDPS+Rx - Adjusting payments with Rx
  7. DxCG - High cost cases and profiling
  8. ACG - Adjusted Clinical Groups system- Uses ADGs
  9. MARA - More detailed level
  10. SCIO - Clinical classification software
  11. Risk/Severity - Inpatient regression
  12. WRA - Wakely Risk Assessment model - Transparent additive
  13. EGM - Medicare performance
  14. ETG - Episodes of Care
  15. MEG - Medicare Episode Groups - Medicare, disease staging
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Useful dimensions of the MARA model

A
  1. Comprehensive set of risk scores by service category
  2. IP and ER scores correlate strongly with the probability of admission and ER events
  3. Individual condition profiles
  4. Recency of care
  5. Persistency of care
  6. Risk drivers
  7. Chronic and non-chronic mapping of each condition group
  8. Identification of issues related to frailty
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Core principles of episode grouper models

A

Followed for the Truven Medical Episode Groups model

  1. Groups all care for one condition, one patient
  2. Describes diagnosed condition, not treated condition
  3. Levels of severity
  4. Ability for diagnosis to evolve within one episode of care
  5. Classifications that are clinically meaningful to providers
  6. Comprehensive and transparent
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Types of drug grouper models

A
  1. Therapeutic class groupers - they group drugs into a hierarchy of therapeutic classes. Examples are American Hospital Formulary Service and Generic Product Identifier
  2. Drug-based risk adjustment models - they infer the member’s diagnosis from the therapeutic class of drugs the member uses, and generate a relative risk score. Examples are Medicaid Rx, Pharmacy Risk Groups, and RxGroups (DxCG).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Common features of Medicare prospective payment systems

A
  1. System of averages
  2. Increased complexity
  3. Relative weights
  4. Conversion factor
  5. Outliers
  6. Updates
  7. Access and Quality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Challenges with patient classification systems based on coding systems

A
  1. Need for new DRGs
  2. ICD Coding
  3. Upcoding
  4. New coding systems
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Goals of risk adjustment for the Arizona Medicaid program

A
  1. Align payment with relative health risk of members at each health plan
  2. Be accurate and unbiased
    1. Accurate means relatively high correlation between projected and actual costs
    2. Unbiased means methodology should not over-compensate for some risk factors at the expense of others
  3. Be as simple as possible while accomplishing these goals
  4. Minimize the administrative burden of developing and implementing the methodology
  5. Be budget neutral
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Medicaid costs that were excluded from weights in calibrated model (reimbursed in other ways)

A
  1. Prior Period Coverage (PPC)
  2. Behavior Health covered by Arizona Dept of Health Services
  3. Costs above reinsurance thresholds
  4. Children’s rehab services
  5. Maternity costs covered by delivery supplement
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Medicaid groups that will NOT have claims based risk adjustment

A
  1. Reconciled risk groups - actual claims used to determine reimbursement
  2. Delivery supplemental rates - case rates paid for deliveries
  3. Option 1 and 2 transplant members - case rates paid for transplants
  4. SOBRA family planning - supplemental payments for women eligible for family planning services but not other Medicaid benefits
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Categories for risk adjustment in MA

A

For existing enrollees:

  1. Community, non-dual eligible, aged
  2. Community, non-dual eligible, disabled
  3. Community, dual eligible - full benefits, aged
  4. Community, dual eligible - full benefits, disabled
  5. Community, dual eligible - partial benefits, aged
  6. Community, dual eligible - partial benefits, disabled
  7. Institutional

For new enrollees:

  1. Non-Medicaid and not originally disabled
  2. Medicaid and not originally disabled
  3. Non-Medicaid and originally disabled
  4. Medicaid and originally disabled
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Experience items to be included in the Medicare BPT

A
  1. Average population risk score
  2. Enrollment
  3. Revenue
  4. Claims
  5. Non-benefit expense
  6. Profit
20
Q

Ways in which the ACA addresses the antiselection and instability created by the guarantee issue requirement

A
  1. Subsidies for applicants with limited income
  2. Employer and individual mandates
  3. Risk adjustment to transfer revenue from plans with low-risk populations to plans with high risk
21
Q

Rating factors allowed under the ACA

A
  1. Age (3:1 max ratio)
  2. Geographic location
  3. Family size
  4. Tobacco use (1.5:1 max ratio)
22
Q

Reasons ACA used concurrent model for risk adjustment

A
  1. First year of ACA didn’t have historical data for prospective calculation
  2. Prospecitve models are less accurate than concurrent models
  3. Churn rates of members through exchanges means that many plans wouldn’t have claims data on members
23
Q

ACA risk transfer formula

A

24
Q

Practical issues with applying risk adjustment models

A
  1. Risk transfer models assume that risk and cost are linearly correlated
  2. MedPac identified these issues with Medicare HCCs:
    1. Only 70 HCCs for risk scoring
    2. Considerable variation within HCCs of patient severity
    3. Doesn’t account for certain racial groups and income levels likely being higher utilizers
    4. Model only uses 1 year of data, may miss chronic conditions
    5. Does not include a factor for the number of conditions
25
Q

Problems with the Massachusetts risk adjustment model

A
  1. Applies to gross premium, not cost of insurance or pure premium
  2. Bias against zero-condition members
  3. Bias against limited network and other low cost plans
  4. Risk adjustment operates on state level instead of regional
26
Q

Potential sources of bias in risk adjustment transfers (ACA)

A
  1. Partial year enrollment
  2. Lack of historical data
  3. Only a fraction of members trigger conditions
  4. High-cost cases - costs and risk score don’t line up
  5. Prospective vs. concurrent models
  6. Market-share
27
Q

Key learnings from DM programs that ACOs should apply to be successful

A
  1. High quality data analytics
  2. Emphasis on EMR/EHR
  3. Importance of economics
    1. Difficult to change patient behavior in a way that produces a measurable financial outcome
    2. Focus on patients with greatest opportunity for cost reduction
  4. Importance of planning and understanding the opportunity
28
Q

Basic CMS beneficiary assignment process

A
  1. Determine ACO cohorts using TIN
  2. ACO submits and certifies participant list
  3. Patients assessed against list of participating professionals to determine if the ACO has plurality of primary care services
29
Q

Criteria for beneficiary to be assigned to a participating ACO

A
  1. Record of Medicare enrollment
  2. At least one month of Part A and B enrollment (no months of A or B only)
  3. No months of Medicare group private enrollment (FFS plans only)
  4. Assigned to only one Medicare shared savings initiative
  5. Lives in US state, territory, or province
  6. Have a primary care service with physician at the ACO
  7. Receive largest share of primary care services from the participating ACO
30
Q

Medicare eligibility enrollment types for separate expenditure calculations for ACO beneficiaries

A
  1. ESRD
  2. Disabled
  3. Aged/dual-eligible Medicare and Medicaid
  4. Aged/non dual-eligible (eligible for Medicare, not Medicaid)
31
Q

ACO risk adjustment - blending risk scores of new and existing members to create single risk ratio

A
  1. Subpopulation & risk adjustment
    1. Continuously-assigned - CMS-HCC claims-based risk adjustment (when avg risk ratio < 1)
    2. Continuously-assigned - Demographic risk adjustment (when avg risk ratio > 1)
    3. Newly-assigned - CMS-HCC risk adjustment (always)
  2. Create a single risk ratio, weighted by relative membership of new and continuous members
32
Q

Adaptations made to CMS-HCCs to develop HHS-HCCs

A
  1. Prediction year (prospective vs concurrent)
  2. Population (aged and disabled vs private individual and small group)
  3. Type of spending (medical only vs medical+drug)
33
Q

Criteria for including HCCs in the HHS risk adjustment model

A
  1. Represent clinically significant, well defined, and costly medical conditions that are likely to be diagnosed, coded, and treated if they are present
  2. Are not especially subject to discretionary diagnostic coding or diagnostic discovery
  3. Do not primarily represent poor quality or avoidable complications of medical care
  4. Identify chronic, predicatble, or other conditions that are subject to insurer risk selection, risk segmentation, or provider network selection
34
Q

Options for improving the HHS-HCC risk adjustment methodology

A
  1. Improve accuracy for partial year enrollees
  2. Use drug utilization
  3. Pooling of high cost enrollees
  4. Evaluate concurrent vs prospective risk adjustment models
35
Q

Benefits and concerns of adding prescription drug utilization to the HHS-HCC risk adjustment model

A
  1. Benefits
    1. Imputing missing diagnoses
    2. Severity indicator for a specific diagnosis
    3. More timely, standardized data
    4. Mitigates financial disincentive to prescribe expensive medications
  2. Concerns
    1. Finding a model
    2. Gaming, perverse incentives, and discretionary prescribing
    3. Sensitivity of risk adjustment to variations in prescription drug utilization
    4. Added administrative burden, complexity, and costs
    5. Availability of outpation drug data only
    6. Multiple indications for most drugs
36
Q

Criteria for selecting drug-diagnosis pairs for a hybrid model

A
  1. Drugs with pattern of non-discretionary prescribing
  2. Avoid drugs with incentives for over-prescribing
  3. Avoid drugs with variations in prescribing across providers, areas, etc
  4. Carefully consider high-cost drugs
  5. Avoid drugs indicated for multiple diagnoses
  6. Avoid drugs indicated for diagnoses not included in HHS-HCC model
  7. Carefully consider drugs in areas with rapid rate of technological changes
37
Q

Criteria for evaluating hybrid risk adjustment models

A
  1. Clinical/face validity
  2. Empirical/predictive validity
  3. Incentives for prescription drug utilization
  4. Sensitivity to variations in prescription drug utilization
  5. Incentives for diagnosis reporting
38
Q

Necessary items for stable and a sustainable individual health insurance market

A
  1. Individual enrollment at sufficient levels and a balanced risk pool
  2. Stable regulatory environment that facilitates fair competition
  3. Sufficient insurer participation and plan offerings
  4. Slow spending growth and high quality of care
39
Q

Key factors for a healthy balanced risk pool with limited market selection

A
  1. Risk stabilization programs (3 R’s)
  2. Outreach and advertising
  3. Medicaid expansion
  4. Ability to develop adequate rates
40
Q

Changes made to the ACA risk adjustment program

A
  1. Duration impact - 2017 - adjustment added for partial year enrollees
  2. Administrative load - 2018 - admin load reduced by 14%
  3. Inclusion of pharmacy data - 2018
  4. Updated weights - 2017/2018
  5. Large claim pooling mechanism - 2018
41
Q

ASOP 12 - Considerations for testing the risk classification system

A
  1. Test long-term viability of the system
  2. Effect of adverse selection
  3. Risk classes used for testing
  4. Effect of changes
  5. Quantitative analysis
42
Q

ASOP 12 - Considerations in selection of risk characteristics to use in a risk classification system

A
  1. Relationship of risk characteristics and expected outcomes
  2. Causality
  3. Objectivity
  4. Practicality
  5. Applicable law
  6. Industry practices
  7. Business practices
43
Q

ASOP 12 - Considerations when establishing risk classes

A
  1. Intended use
  2. Actuarial considerations
    1. Adverse selection
    2. Credibility
    3. Practicality
  3. Other Considerations - legal compliance, industry practice, business practice limitations
  4. Reasonableness of results
44
Q

ASOP 45 - Considerations when selecting a risk adjustment model

A
  1. Transparency
  2. Pratical considerations
  3. Predictive ability
  4. Model version
  5. Reliance on experts
  6. Impact on program
  7. Timing of data collection, measurement, and estimation
  8. Intended use
  9. Population and program

HINT: T PAVE IT UP

45
Q

ASOP 45 - Considerations regarding consistency of input data

A
  1. Provider contracts
  2. Diagnostic services
  3. Coding and other data issues

Specialty (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions