Objective 4 - Risk Adjustment

Question 1

Reasons for reinsuring accident and health products (3; GHRM-113-23)

Answer 1

1) To transfer risk (the primary reason)
2) To enable an insurer to offer products in a specific market in which it lacks expertise. This allows the insurer to provide a broad range of products.
3) To share the financial load. This is sometimes done for products that require large amounts of capital, such as individual LTC.

Question 2

Proportional reinsurance methods (4; GHRM-113-23)

Answer 2

1) Coinsurance - the reinsurer accepts the ceding company’s reinsurance premium and pays its share of benefits. The reinsurer pays a ceding allowance to cover a portion of the ceding company’s commission and expenses. Approaches for establishing risk sharing are:
a) Fixed or excess share - the ceding company retains a fixed, level amount of risk (making this a non-proportional method)
b) Quota share - the ceding company retains a fixed percentage of each risk

2) Modified coinsurance - works just like coinsurance, except the assets backing the reserves are held by the ceding company

3) Funds withheld coinsurance - ceding company keeps assets and reinsurer gets credit on accounting books for the receivable

4) Risk premium reinsurance - annual premium is based on anticipated claim cost for given issue age, policy duration, and underwriting class

Question 3

Nonproportional reinsurance methods (5; GHRM-113-23)

Answer 3

1) Extended wait (aka extended elimination period or extended deductible) - reinsurance benefits begin only after the claims have reached a specified duration or amount

2) Excess reinsurance or stop loss reinsurance - provides coverage for claims in excess of a defined level
a) Individual excess or specific stop loss - provides payment if the total benefits on a single individual exceed a specified attachment point or deductible
b) Aggregate stop loss - provides a benefit if total retained claims on the entire group or portfolio exceed a specified attachment point, typically defined as a percentage (such as 125%) of expected claims

3) Specified benefits or carve out benefits - for medical insurance, certain benefits may be carved out, such as premature births, organ transplants, and trauma. Disability income policies occasionally carve out accidental benefits.

4) Claim takeover reinsurance and runoff blocks - the reinsurer assumes the risk (and control) on future runoff of a known block of claims

5) Catastrophe covers - provide coverage in the event of a catastrophe

Question 4

Decisions to make when setting retention limits for disability income insurance (2; GHRM-113-23)

Answer 4

1) The method of determining the retention - the insurer may retain a maximum amount per month or it may set a limit on the total benefit paid for the entire benefit period
2) The amount of the maximum claim - the insurer might set limits for disability income that would produce present values of benefits approximately the same as the maximum retention for life insurance

Question 5

Purchasers of medical reinsurance (4; GHRM-113-23)

Answer 5

1) Traditional insurers offering both first dollar insurance and excess of loss coverage
2) Employers providing self-insured benefits to employees
3) HMOs providing services
4) Certain providers that offer prepaid benefit plans

Question 6

Approaches for defining coverage periods for health reinsurance (2; GHRM-113-23)

Answer 6

1) Losses Occurring Basis - claims are only covered if they occur during the agreement year, regardless of the effective date of the risks accepted by the insurer. Is used most commonly for excess arrangements.
2) Risk Attaching Basis - reinsurance period for underlying risk coincides with the insurer’s policy year. Commonly used for proportional reinsurance.

Question 7

Primary approaches for reinsurance of major medical policies (5; GHRM-113-23)

Answer 7

Note: Coinsurance is rarely used today, but stop loss is very common

1) Quota share coinsurance - fixed percentage of all policies is reinsured
2) Specific stop loss (or excess) - ceding company retains all claims below the attachment point
3) Aggregate stop loss (or excess) - Usually not for 100% of excess claims, since ceding company required to keep portion of risk to ensure appropriate claims handling
4) Combined specific and aggregate stop loss - can be combined into a single package, where the cost of one impacts the other
5) Carve out coverages - usually covers 100% of cost of benefit

Question 8

Key questions regarding the source of business for medical reinsurance (6; GHRM-113-23)

Answer 8

1) Is it coming from an HMO?
2) Does the plan include PPO networks?
3) How do benefits vary inside and outside the network?
4) How are reasonable and customary limits applied?
5) What employer groups are targeted and how are occupational hazards handled?
6) What amounts are self-insured employer groups required to retain?

Question 9

Distinct markets of medical reinsurance (4; GHRM-113-23)

Answer 9

1) Smaller insurers - reinsurer provides expertise and oversight
2) Large writers - more sophisticated and get coverage for high deductible amounts; decision of reinsurer made mostly on price
3) Self-funded employers - reinsurers need strong underwriting staff, must understand TPA claim practices, admin. capabilities, and reputation
4) HMOs - requires special expertise

Question 10

Uses of reinsurance for medical coverages other than major medical (7; GHRM-113-23)

Answer 10

1) Fixed benefits medical and surgical products - reinsurance is rarely purchased for claims protection, but quota share coinsurance may be purchased to support growth
2) Specific or dread disease products - provide payment upon diagnosis; reinsurance is rarely used since most insurers retain all the risk on these policies
3) Accident and AD&D coverages - except for very large coverages, reinsurance is rarely used. Some reinsurers offer coverage for accidents on a portfolio basis, similar to catastrophe reinsurance.
4) Medicare Supplement - reinsurance is attractive for an insurer that no longer wants to retain the risk or manage the product. Quota share coinsurance is sometimes used because the product is relatively capital intensive. Fronting is popular - marketing org. sells product through another insurer (front) and most or all business is ceded to the first insurer for a fee.
5) Critical illness - coinsurance has been used, with the reinsurers providing expertise and product design advice
6) Dental and vision coverages - many group insurance plans offer these coverages and then use coinsurance to transfer the risk to reinsurers that specialize in these coverages
7) Other uses of reinsurance include captive reinsurers for employee benefits, stop loss for providers, securitizations of health insurance, and capital relief provided by a portfolio reinsurance agreement

Question 11

Impacts of the ACA on the reinsurance market (5; GHRM-113-23)

Answer 11

1) The requirement for unlimited benefits and the inclusion of certain mandatory benefits led to cost adjustments
2) Reinsurance of limited benefit medical policies is now seldom needed
3) Reinsurance has been provided to plans on the ACA exchanges
4) There has been little effect on the reinsurance of medical benefits, other than to increase demand
5) Annual and lifetime benefit limits for major medical plans were removed. Reinsurance solutions to this are:
a) An insurer may purchase separate layers of stop loss reinsurance. Each layer other than the top layer would have limits.
b) The insurer could retain the excess risk (have no unlimited layer)
c) A reinsurer may accept the unlimited risk, but then purchase unlimited retrocession coverage.

Question 12

Steps for implementing risk adjustment into a Medicaid managed care program (9; Duncan 13)

Answer 12

1) Decide which risk adjustment system will be used - there are various commercially available models. Should choose a system based on the data used and the ability to customize it.
2) Decide what types of data should be used in the risk adjustment system - includes demographic information and claims and pharmacy data.
3) Decide which Medicaid eligibility groups will be risk-adjusted and which sub-populations may be excluded
4) Decide whether the risk adjustment system should be prospective (use experience period data to estimate future morbidity) or concurrent (use data from the current period to estimate morbidity for that period)
5) Decide whether to base the risk adjustment factors on the individuals enrolled during the rating period or during the experience period
6) Decide whether to customize the risk weights inherent in the risk adjustment model - may be needed due to differences in the state program as compared to the population used to develop the model
7) Decide on criteria for including individuals in the risk adjustment calculations - many states require at least six months of eligibility exposure
8) Develop criteria for claims records to be included in the risk adjustment model
9) Determine the phase-in schedule and whether or not risk corridors will be used

Question 13

Goals of risk adjustment for the Arizona Medicaid program (5; Duncan 13)

Answer 13

1) Align payment with the relative health risk of members at each health plan
2) Be accurate (relatively high correlation between the projected cost of the population and actual cost) and unbiased (methodology should not overcompensate for some risk factors at the expense of others)
3) Be as simple as possible while accomplishing other goals
4) Minimize the administrative burden of developing and implementing the methodology
5) Be budget neutral

Question 14

Methodology used to develop the Arizona Medicaid risk adjustment model (11; Duncan 13)

Answer 14

1) Model selected - Symmetry’s Episode Risk Groups (ERG) model
2) Type of data used - diagnosis codes and procedural information from medical data and National Drug Codes from pharmacy data
3) Data timing - three months of claim run-out was used
4) Eligibility groups - risk adjustment was applied to prospective, non-reconciled groups
5) Model calibration - the model was recalibrated by developing risk weights through a linear regression model based on Arizona Medicaid data, and then credibility weighting those rates with the model’s original risk weights
6) Geographic issues - risk adjustment will take place at the geographical service area and risk group level
7) Individual approach - risk scores calculated during the experience period will follow the individual during the rating period. This will accurately reflect movement of individuals between health plans.
8) Risk factors are updated once per year
9) Risk factors for new members - members with at least six months of enrollment (“long” cohort) during the experience period will be given a claims-based risk factor. Other members (“short” cohort) will be given a risk factor that is the average of an age-gender factor and an adjusted plan factor. Adjusted plan factor = (average ERG risk score of long cohort / pure age-gender factor of long cohort) * pure age-gender factor of short cohort
10) Phase-in - risk adjustment is being phased in such that only 80% of the 2009 rate is risk adjusted
11) Risk factors for newborns - a different approach is needed because newborns do not have prior year claims from which to develop condition-based risk scores. Claims of the prior cohort of newborns in the experience period are used to project newborn experience in the rating period.

Question 15

Formulas for calculating an MCO’s risk score for the Arizona Medicaid program (5; Duncan 13)

Answer 15

1) Average ERG risk score for long cohort
a) An unadjusted risk score is calculated as the sum over all risk factors and demographics of the risk weights multiplied by frequencies. The frequencies are the portion of the cohort with each risk factor or demographic.
b) For the Transitional Aid to Needy Families group, the final risk score equals the unadjusted risk score divided by a scaling factor

2) Total average risk score = % of members in long cohort * average ERG risk score for long cohort + % of members in short cohort * risk factor for short cohort
a) Risk factor for short cohort = 50% * adjusted plan factor for short cohort + 50% * pure age-gender factor of short cohort

3) The above formulas are calculated for the given MCO and for all MCOs in total. The MCO’s relative risk score = MCO total average risk score / average risk score for all MCOs

4) Relative risk score with phase-in = 80% * relative risk score + 20% * 1.00

5) A budget neutrality adjustment may also be applied to get the final relative risk score

Question 16

Formulas for calculating final capitation rates for the Arizona Medicaid program (3; Duncan 13)

Answer 16

1) Capitation rate to be risk adjusted - base capitation rate - bid risk contingency - bid admin cost - 2% premium tax
2) Risk-adjusted capitation rate (before retention) = capitation rate to be risk adjusted * risk adjustment factor (relative risk score)
3) Final risk-adjusted capitation rate = risk-adjusted capitation rate (before retention) + bid contingency + bid admin cost + 2% premium tax

Question 17

Common hypotheses for the member selection patterns observed in Medicare Advantage plans versus traditional Medicare FFS (2; Duncan 14)

Answer 17

Those enrolling in Medicare Advantage plans have been observed to be materially healthier. Theories for why include:
1) Healthy enrollees are less reluctant to change benefit plans, so they are more likely to sign on with Medicare Advantage
2) Managed care organizations restrict access to certain network health care providers. Since less healthy Medicare enrollees generally have established provider relationships, they are more reluctant to leave traditional Medicare and risk losing access to their preferred providers.

Question 18

Impacts of Medicare risk adjustment on Medicare Advantage Organizations (MAOs) (5; Duncan 14)

Answer 18

1) MAOs are responsible for capturing complete and accurate diagnoses for their members’ health conditions
2) MAOs must validate their risk adjustment data for audits conducted by CMS, referred to as Risk Adjustment Data Validation (RADV) audits
3) Medicare risk model changes can have a greater impact on a MAO than it does on a national basis. MAOs need to conduct their own new model vs old model analysis to estimate the differences.
4) Medicare risk adjustment is a complicated process involving a large amount of operational complexity for MAOs
5) MAOs do not compete on selecting members who are better risks. To be successful, MAOs focus on better outcomes, improved population health, and controlling per capita cost

Question 19

Methodology for calculating member risk scores for Medicare Advantage Part C (6; Duncan 14)

Answer 19

1) A member’s risk score is the sum of weights that reflect that member’s characteristics, including an age/gender score and a health condition score based on the coefficients attached to 79 different health hierarchical condition categories (HCCs)
2) A member may have multiple HCCs and weights are included for each applicable HCC
3) There are also several weights that result from interactions between HCCs
4) Weights are applied hierarchically. If a member has multiple HCCs in the same hierarchy, only the weight for the most severe HCC is counted. The weights of the less-severe HCCs are “trumped”.
5) Condition scores are prospective factors. Diagnoses in the prior year are used to predict Medicare health claim costs in the current year.
6) For members that are new to Medicare, CMS provides only age/gender factors. These factors are higher than those for ongoing beneficiaries because the full responsibility for predicting future cost is assigned to only the age/gender factors.

Question 20

Categories of Medicare Advantage members (11; Duncan 14)

Answer 20

Members are split into these different categories for determining Part C risk scores. Each category has its own age/gender and HCC factors.

Categories for ongoing members:
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

Categories for new enrollees:
8) Non-Medicaid and not originally disabled
9) Medicaid and not originally disabled
10) Non-Medicaid and originally disabled
11) Medicaid and originally disabled

Question 21

Experience items included in the Medicare Bid Pricing Tool (BPT) (6; Duncan 14)

Answer 21

The BPT is an Excel workbook pricing form for each of Part C and D, which CMS requires Medicare Advantage plans to use. The following past experience items are projected forward two years from the base year to the contract year.
1) Average population risk score
2) Enrollment level (in member months)
3) Revenue
4) Claims
5) Non-benefit expense
6) Profit

Question 22

CMS requirements for projected risk scores in the BPT (6; Duncan 14)

Answer 22

The formula for projecting risk scores is established by CMS, but each Medicare Advantage Organization is allowed to develop its own trend rate to use in that projection. The projected risk scores must:
1) Be based on the methodology for calculating the risk scores as discussed in the Rate Announcement
2) Be calculated using the CMS-HCC risk adjustment model
3) Reflect the expected risk score trend at the bid level
4) Be appropriate for the expected population
5) Be adjusted for FFS normalization
6) Include the appropriate Medicare Advantage coding adjustment

Question 23

Considerations for projecting risk scores in the BPT, other than those prescribed by CMS (5; Duncan 14)

Answer 23

1) The expected trend in risk scores from changes such as better diagnosis coding - risk scores tend to drift upwards over time due to improved and more complete coding
2) The risk scores of new entrants
3) Population change must be estimated
4) A mortality factor should be considered, as claims of deceased patients are heavily skewed (average mortality ~5%, claims from these members nearly 1/3 of total)
5) Most MAOs perform longitudinal analysis on their membership stratified into “stayers”, “leavers”, and “joiners”

Question 24

Consequences of an incorrect risk score estimate for Medicare Advantage (2; Duncan 14)

Answer 24

1) If the overstatement is too extreme, it can result in losses from which the MAO may not recover. An overstated risk score improvement may also force the MAO to offer richer benefits than it should
2) Understated projected risk scores may lead to an uncompetitive product

Question 25

Reasons the ACA was enacted (3; Duncan 21)

Answer 25

1) Increase the quality and affordability of health insurance
2) Lower the uninsured rate by expanding public and private insurance coverage
3) Reduce the costs of healthcare for individuals and the government

Question 26

Ways in which the ACA addresses the antiselection and instability created by the guaranteed issue requirement (3; Duncan 21)

Answer 26

1) Subsidies are available for individuals with limited incomes
2) Employers are required to provide insurance, and all residents ineligible for employer coverage must purchase individual coverage or pay a penalty.
3) A risk adjustment mechanism transfers revenue from plans with relatively low-risk populations to plans with relatively high-risk populations. Risk corridors also transferred revenue from profitable to unprofitable plans.

Question 27

Features of the ACA Risk Adjustment program (5; Duncan 21)

Answer 27

1) Payment adjustments sum to zero (revenue neutrality) - lower risk plans pay into reimbursement scheme and higher risk plans are reimbursed from that money
2) Transfers done separately for individual and small group markets
3) Calculations performed separately for each metal level
4) HHS-HCC model used for risk-adjustment (with revisions for population covered) - different than CMS-HCC model used for Medicare
5) Different models used for sub-populations (ages 0-1, 2-20, 21+)

Question 28

Rating factors allowed by the ACA (4; Duncan 21)

Answer 28

1) Age (limited to 3:1 ratio)
2) Geographic rating area
3) Family size
4) Tobacco status (rates can be increased 50% for tobacco users)

Question 29

Reasons why the ACA uses a concurrent risk adjustment model (3; Duncan 21)

Answer 29

1) For the first year of the ACA, most exchange participants were expected to be previously uninsured, so no data was available to perform a prospective calculation.
2) Prospective risk adjustment models are less accurate than concurrent models, as demonstrated in different SOA comparative studies.
3) The churn rate of members through the exchanges has been high, so even in later years many plans will still not have claims data on members

Question 30

ACA Risk Transfer formula (Duncan 21)

Answer 30

Question 31

ACA Risk Transfer formula components (7; Duncan 21)

Answer 31

For plan i,

1) Ti is the transfer amount per billable member month. A positive amount means the plan receives a payment, while a negative amount represents payment by the plan
2) PLRS - plan liability risk score (AV + plan’s enrollee health status risk)
2) IDF - induced demand factor
3) GCF - geographic cost factor
4) AV - actuarial value (associated with plan’s metal level)
5) ARF - allowable rating factor (average of rating factors)
6) s - plan enrollment market share
7) P - market-wide average premium

Question 32

Practical issues with applying risk adjustment models (3; Duncan 21)

Answer 32

1) Risk transfer models generally assume that risk and cost are correlated, so a 1% increase in risk is assumed to increase costs by 1%. However, not all cost-risk relationships are linear. As a result, these models overcompensate some plans and undercompensate other plans.
2) Various issues related to HCCs identified by the Medicare Payment Advisory Commission (MedPac) (separate list)
3) Several issues exists in ACA risk adjustment (separate list)

Question 33

Issues in HCCs related to risk adjustment (5; Duncan 21)

Answer 33

1) Although the CMS-HCC risk adjusters map diagnosis codes to 189 HCCs, only 70 HCCs are actually used for risk scoring. Some fairly prevalent medical conditions are not accounted for.
2) There is considerable variation within HCCs in terms of patient severity and experience
3) Certain racial groups and income levels are likely to be higher consumers of healthcare, but this is not reflected in the model
4) Because the model only uses one year of data for determining risk scores, the model under-predicts for some chronic conditions where patients may not have a claim that year
5) The standard model does not include a factor for the number of conditions, but MedPAC has found that this factor would lead to more accurate predictions

Question 34

Problems with the Massachusetts risk adjustment model (4; Duncan 21)

Answer 34

1) Risk adjustment applies to the gross premium, not the cost of insurance or pure premium, so transfers include part of the expense margin.
2) The model has been shown to be biased against zero-condition members, particularly at the younger ages
3) There is also bias against limited network and other lower cost plans. Risk transfers have been observed to exceed net income for some of these plans.
4) Risk adjustment operates at the state, rather than regional level, creating another bias

Question 35

Issues and potential improvements for the national ACA risk adjustment model (6; Duncan 21)

Answer 35

1) The model is not accurate for adults with partial year enrollment
2) Lack of historical data - the ACA uses only one year of claim data in a concurrent model, which fails to properly reflect the risk of chronic members who may not have a claim in some years
3) Only a fraction of members trigger conditions - this could be because a provider fails to code a condition or because the condition present is not mapped to an HCC
4) Because risk scores do not track costs well at the extremes, high-risk members may experience costs that are disproportionate to their risk scores
5) Prospective vs. concurrent models - sufficient data may now exist to move to a prospective model, but CMS has rejected making this change because it believes the concurrent model is best for this population
6) Market-share - insurers with small market shares and with premiums that are much different than the statewide average are likely to see revenue transfers that are unrelated to their own premiums

Question 36

Notable changes in the ACA risk adjustment model over time (9; Past and Future of ACA Risk Adjustment)

Answer 36

1) Annual coefficient recalibration
2) 2015 - transition to model using condition categories based on ICD10 codes
3) 2017 - added duration factors to adjust for partial year enrollees, improving profitability by member duration
4) 2018 - added prescription drug classes to better account for claims costs
5) 2018 - reduction of statewide average premium (14% administrative load no longer included) and change transfers to paid claims basis
6) 2018 - added high-risk pool (HCRP) sharing (60% share above a $1M threshold)
7) 2018 - adjustments from prior year audits
8) 2019 - phase-in of External Data Gathering Environment (EDGE) data in coefficient calibration
9) 2021 - updated condition category calibration - should more accurately capture relative costs by medical condition

Question 37

Impacts on insurers from ACA risk adjustment changes over time (7; Changing with the Times)

Answer 37

Model impacts - includes only the model changes over time, the population is held fixed
1) The “condition” component of the risk score (HCC plus RxC) is an increasing proportion of the total
2) Composite risk scores are decreasing

Market impacts - impacts include annual model changes as well as enrollment changes
3) Risk adjustment represents a large portion of ACA premium
4) Variability of risk adjustment as a % of premium has remained high in both individual and small group markets
5) Risk adjustment shows some stability at the market level, but it is very much an issuer-specific experience
6) Many insurers do experience large swings in ACA transfers every year which is difficult to account for when setting premium rates. Up to 30% of insurers reverse position (from receivable to charge or vice versa) from the prior year.
7) The risk adjustment transfer payment approach is sensitive to both enrollment count (insurer size) and the change in enrollment counts (market shifts)

Question 38

Further areas for improvement in ACA risk adjustment model (8; Changing with the Times)

Answer 38

CHINA PGV

1) Further develop coefficients (larger portion of EDGE data and recent market changes)
2) Change HCC/RxC values and categories (leverage precision of ICD10 codes)
3) Refresh CSR IU factors
4) Use non-linear model for calibration process
5) Use additional factors (issuer network characteristics or premium levels)
6) Use other predictive factors (social determinants of health and socioeconomic data)
7) Update governance procedures (allow more up-to-date information or time for issuers to understand changes)
8) Enhance risk adjustment data validation (better align ultimate risk transfer)

Question 39

Options for improving the HHS-HCC risk adjustment methodology (4; GHRM-112-23)

Answer 39

1) Improve the accuracy of the model for partial year enrollees
a) Length of enrollment could be included as a new indicator variable
b) Separate models could be produced for different enrollment period groups (months 1-4, 5-8, 9-12)
c) Second approach appears to predict more accurately, but it may present false precision when predicting costs for conditions with small sample sizes and it adds to the complexity of the risk adjustment methodology, which already includes separate models by age and metal level

2) Use prescription drug utilization as a predictor in the model
3) Pooling of high cost enrollees
4) Evaluating concurrent and prospective risk adjustment models

Question 40

Benefits of adding prescription drug utilization to the HHS-HCC risk adjustment model (4; GHRM-112-23)

Answer 40

1) Imputing missing diagnoses - drug utilization data may capture the existence of some conditions that are missing in diagnoses entered on medical claims, particularly for chronic conditions
2) Severity indicator for a specific diagnosis - the presence of certain drugs can indicate the severity of illness for some HCCs
3) More timely, standardized data - drug data can be available more quickly than medical claim data, is often more complete, is often easier to access, and is more standardized because it does not vary with provider coding patterns
4) Mitigates the financial disincentive to prescribe expensive medications - a risk adjustment model that incorporates prescription drug utilization will compensate plans that cover high-cost medications, reducing the incentive for plans to restrict access to these medications

Question 41

Concerns about adding prescription drug utilization to the HHS-HCC risk adjustment model (6; GHRM-112-23)

Answer 41

1) Risk adjustment models that use drug information are not as common as models based only on medical information, so they are not as well understood or accepted
2) Gaming, perverse incentives, and discretionary prescribing - gaming occurs when a drug is prescribed in order to trigger a higher payment. Drug models are particularly susceptible to gaming because some relatively low-cost drugs are linked to high medical costs. Financial incentives may inappropriately influence treatment decisions.
3) Sensitivity of risk adjustment to variations in prescription drug utilization - many factors other than health status affect drug utilization, and risk adjustment based on drug information will reflect these factors
4) Added administrative burden (to calibrate and apply the model), operational complexity, and costs (due to data reporting requirements and frequent updates)
5) Availability of outpatient drug data only - some drug models omit drugs provided in a hospital setting, which may make hospitalized patients appear to be severely less ill
6) Multiple indications for most drugs - many drug classes are widely prescribed “off-label” for indications that are not FDA-approved, so utilization of these drug classes does not always indicate the presence of a specific diagnosis

Question 42

Factors other than health status that affect drug utilization (5; GHRM-112-23)

Answer 42

1) Plan and physician prescribing patterns
2) Cost sharing features - induced utilization
3) Drug utilization management features
4) Proclivities of providers for using drug versus non-drug treatments for a medical condition
5) The income level of enrollees

Question 43

Criteria for evaluating hybrid risk adjustment models (5; GHRM-112-23)

Answer 43

“Hybrid” models are those that incorporate both diagnoses and prescription drugs

1) Clinical face validity - should be clinically valid in the relationship between the risk markers (diagnoses and drugs) and health care expenditures, and in the relationship between drugs and associated diagnoses
2) Empirical / predictive accuracy - drugs added to the model should increase the model’s accuracy in predicting health expenditures
3) Incentives for prescription drug utilization - adding drugs should be done in a way that minimizes incentives for over-prescription of drugs to maximize risk transfers, but does not discourage needed drugs
4) Sensitivity to variations in prescription drug utilization - should incorporate variations in drug utilization that measure differences in enrollee health status, not variation due to other factors
5) Incentives for diagnosis reporting - accurate and complete diagnosis reporting should not be discouraged by reducing predicted expenditures when additional diagnoses are appropriately reported

Question 44

Approaches for adding prescription drug utilization to a risk adjustment model (5; GHRM-112-23)

Answer 44

1) Statistical predictive power approach - drug classes are included in the model on purely statistical grounds. Advantage: this approach allows for a linkage between a drug and poor health in general. Disadvantage: by omitting clinical considerations it makes interpretation of model coefficients difficult and leads to less clinical face validity.

Following approaches are all conceptual
2) Imputation - using drug data to impute missing diagnoses. The predicted incremental cost is the same regardless of how a health condition is identified, whether by a drug indicator only, a diagnosis indicator only, or both indicators.
3) Severity - using drug data as a severity indicator for a specific diagnosis. Only if the drug class and a specific diagnosis are present will the model predict incremental costs beyond the diagnosis alone.
4) Rx dominant - individuals taking a drug are assumed to be more severely ill (have higher projected costs) than individuals not taking the drug who only have the associated diagnosis marker
5) Flexible, generalized, empirical framework - each drug-diagnosis pair enters the model with three indicator variables: a diagnosis indicator, a drug class indicator, and an interaction indicator. Each indicator has a coefficient that predicts the incremental costs for that indicator.

Question 45

Criteria for selecting drug-diagnosis pairs for a hybrid model (7; GHRM-112-23)

Answer 45

1) Select drugs with patterns of non-discretionary prescribing
2) Avoid drugs where there are incentives for over-prescribing
3) Avoid drugs where there are variations in prescribing across providers, practices, and areas
4) Carefully consider selection of high-cost drugs. In some cases, including the drug in the model may reduce the incentives for insurers to strive for greater efficiency.
5) Avoid drugs indicated for multiple diagnoses
6) Avoid drugs indicated for diagnoses not included in the HHS-HCC model
7) Carefully consider selection of drugs in an area exhibiting a rapid rate of technological change (which could make cost predictions inaccurate when based on previous years of data)

Question 46

HHS considerations when selecting drug-diagnosis pairs to include in a hybrid HHS-HCC model (7; GHRM-112-23)

Answer 46

1) Empirical considerations - a wide range of exploratory data analysis was performed to determine pairs to consider. Then stepwise regression was used to determine which drug classes added the most predictive power to the existing HHS-HCC model.
2) Clinical considerations - doctors and pharmacists were consulted to provide deeper insights into the medical links between health conditions and the drug groups being considered, and to identify the potential for gaming for each drug being considered
3) Imposing medical restrictions based on days’ supply or number of prescriptions in order to trigger a drug indication
4) Whether to split certain drug classes or restrict a drug-diagnosis interaction to certain drugs within a class
5) HHS examined different models that include imputation-only versus imputation and severity approaches
6) Prophylactic (preventive) use of drugs - drugs are sometimes used in persons at risk of disease but who do not actually have the disease
7) Multiple indications for drugs - drug classes are often indicated for multiple diagnoses

Question 47

Market forces that led to a stable and sustainable individual insurance market (7; Creating Stability in Unstable Times)

Answer 47

Stable and sustainable market forces
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 to provide insurer competition and consumer choice
4) Slow spending growth and high quality of care

Unstable and unsustainable market forces
5) Uninsured rates too high, with not enough young and healthy enrollees
6) Issuers leaving the market
7) Consumers have fewer choices for richer and wider network plans

Question 48

Key factors for a healthy balanced risk pool with limited market selection (7; Creating Stability in Unstable Times)

Answer 48

1) The risk stabilization programs - risk adjustment, reinsurance, and risk corridors
2) Outreach and advertising - these were especially effective for individuals eligible for subsidies
3) Medicaid expansion - a study found that expansion was correlated with lower individual premiums
4) Ability to develop adequate rates - requires a stable regulatory environment and issuers need data/knowledge of the risk pool

To encourage young and healthy enrollment:
5) The individual insurance mandate
6) Subsidies for individuals
7) An open enrollment period and limited special enrollment periods (SEPs)

Question 49

Results from the 2015 individual market study (7; Creating Stability in Unstable Times)

Answer 49

1) Lean metal level plans were profitable and rich plans were not (issuers mostly stopped offering platinum)
2) The richer silver cost-sharing reduction (CSR) plans were profitable (due to CSR IU factors being too high)
3) PPO plans were less profitable than HMO plans
4) After risk adjustment, sicker patients are not necessarily driving losses
5) After risk adjustment, younger demographics tend to be less profitable (healthier, less conditions to trigger RA transfers)
6) Partial year enrollees and SEP enrollees are relatively less profitable than full year enrollees (improved in 2017 methodology changes)
7) Results were different by market, with small group showing profitability in the richer metal level plans

Question 50

State stabilization attempts to bend the cost curve and lower premiums (6; Creating Stability in Unstable Times)

Answer 50

1) Change in age factors or subsidy structure to encourage younger individuals to enroll
2) Alter rating areas - encourage issuer participation in at-risk or bare counties
3) Require Medicaid MCOs to participate in individual market (as a condition for Medicaid participation)
4) State-based option where state operates a plan in the market
5) Medicaid buy-in program - individuals can purchase Medicaid coverage
6) Programs impacting overall cost and quality to try to “bend the trend” (e.g. value based care)

Many of these approaches would first require changes at the federal level

Question 51

Actuarial standards for the use of data (7; ASOP #23 - Data Quality)

Answer 51

1) Data that is completely accurate, appropriate, and comprehensive is frequently not available, so the actuary should use available data that allows the actuary to perform the analysis
2) Considerations in selecting data (separate list)
3) Review of data - the actuary should review the data for reasonableness, unless such a review is not necessary or practical
4) The actuary should use appropriate data (separate list)
5) Reliance on data and other information supplied by others - the accuracy of this information is the responsibility of those who supply it. The actuary may rely on this information, but should disclose this reliance.
6) Confidentiality - the actuary should handle data containing confidential information consistent with Precept 9 of the Code of Professional Conduct
7) Limitation of the actuary’s responsibility - the actuary is not required to audit the data or determine whether data supplied by others is intentionally misleading

Question 52

Considerations in selecting data to use in an actuarial analysis (9; ASOP #23 - Data Quality)

Answer 52

1) The scope of the assignment and the intended use of the analysis
2) The desired data elements and possible alternative data elements
3) Whether the data is appropriate and sufficiently current
4) Whether the data is internally consistent
5) Whether the data is reasonable given relevant external information that is readily available
6) The degree to which the data is sufficient for the analysis
7) Any known significant limitations of the data
8) The availability of alternative data, and the benefit and practicality of obtaining this data
9) Sampling methods that were used to collect the data

Question 53

Categories of appropriateness of data used in an actuarial analysis (5; ASOP #23 - Data Quality)

Answer 53

1) The data is of acceptable quality to perform the analysis
2) The data requires enhancement before the analysis can be performed, and it is practical to obtain additional or corrected data
3) Judgmental adjustments or assumptions can be applied to the data, or the analysis results, to allow the actuary to perform the analysis
4) The data is likely to have significant defects
5) The data is so inadequate that it cannot be used to satisfy the purpose of the assignment

Question 54

Required documentation related to data quality (9; ASOP #23 - Data Quality)

Answer 54

1) The source of the data
2) Any limitations on the use of the actuarial work product due to uncertainty about data quality
3) Whether the actuary reviewed the data, and any limitations due to data that was not reviewed
4) A summary of unresolved concerns the actuary may have about questionable data values
5) A summary of significant steps the actuary has taken to improve the data
6) A summary of significant judgmental adjustments or assumptions the actuary applied to the data or to the results
7) The existence of results that are highly uncertain or potentially biased due to the quality of the data
8) The extent of the actuary’s reliance on data and other information supplied by others
9) Disclosures in accordance with ASOP #41 if:
a) Any material assumption or method was prescribed by law
b) The actuary relies on other sources and thereby disclaims responsibility for any material assumption or method
c) The actuary has otherwise deviated materially from the guidance of this ASOP

Question 55

Recommended practices for actuarial communications (4; ASOP #41 - Actuarial Communications)

Answer 55

1) Actuarial communications should meet the following requirements:
a) The form and content of the communication must be appropriate for the given circumstances
b) The communication should be clear
c) Each communication should be issued within a reasonable time period
d) All actuaries responsible for the communication should be clearly identified

2) The actuary should complete an actuarial report if the actuary intends the findings to be relied upon by any intended user
3) Some circumstances (such as regulations) may constrain the content of an actuarial report. In these cases, the actuary should follow the guidance of this standard to the extent reasonably possible.
4) The actuary should recognize the risk of unintended users misusing an actuarial document, and should take reasonable steps to ensure it is clear and presented fairly

Question 56

Disclosures required in an actuarial report (10; ASOP #41 - Actuarial Communications)

Answer 56

This report states the actuarial findings and identifies the methods, procedures, assumptions, and data used

1) The intended users of the report
2) The scope and intended purpose of the assignment
3) The acknowledgement of qualification as specified in the Qualification Standards
4) Any cautions about risk and uncertainty
5) Any limitations or constraints on the use or applicability of the findings
6) Any conflict of interest
7) Any information on which the actuary relied that has a material impact on the findings and for which the actuary does not assume responsibility
8) The information date (date through which data and other information has been considered)
9) Subsequent events (may have a material effect on the actuarial findings)
10) If appropriate, the documents comprising the actuarial report

Question 57

Disclosure requirements for assumptions and methods used in an actuarial report (3; ASOP #41 - Actuarial Communications)

Answer 57

1) The communication should identify the party responsible for each material assumption and method
2) If the assumption or method is prescribed by law, disclose the applicable law, the assumptions or methods affected, and that the report was prepared in accordance with the law
3) If a material assumption or method is selected by another party, the actuary has three choices:
a) If it does not conflict with the actuary’s professional judgment, no disclosure is needed
b) If it significantly conflicts with the actuary’s professional judgment, then disclose this fact
c) If the actuary is unable or not qualified to judge its reasonableness, then disclose this fact

In the case of either b or c, also disclose the affected assumption or method, the party who set it, and the reason it was set by this party, rather than the actuary

Question 58

Considerations when selecting a risk adjustment model per ASOP #45 (9; ASOP #45 - The Use of Health Status Based Risk Adjustment Methodologies)

Answer 58

1) Intended use - consider the degree to which the model was designed to estimate what the actuary is trying to measure
2) Impact on program - consider whether the risk adjustment system may cause changes in behavior because of underlying incentives
3) Model version - if a new version of a previously-utilized model is used, consider the materiality of changes to the model
4) Population and program - consider if the population and program to which the model is being applied are consistent with those used to develop the model
5) Timing of data collection, measurement, and estimation - consider the impact of timing differences between when the model is developed and when it is applied
6) Transparency - consider whether the model provides an appropriate level of transparency for the intended use
7) Predictive ability - consider the predictive ability of the model and the characteristics of the various common predictive performance measures
8) Reliance on experts - consider whether the individuals incorporating their specialized knowledge into the model are experts in risk adjustment
9) Practical considerations - consider practical limitations, such as the cost of the model, the actuary’s familiarity with the model, and its availability