Inner TRIM3 Masthead
TRIM3 Logo

MedicaidCHIP version 31

Version History

Medicaid provides health insurance to low-income individuals. The groups primarily served are the elderly, the disabled, and families with children, although other adults may also be covered. States may extend additional coverage to children and pregnant women via CHIP.

TRIM3's MedicaidCHIP module simulates eligibility for Medicaid and Children's Health Insurance Programs (CHIP) and identifies which eligible individuals are actually enrolled in the program. Eligibility is simulated on a monthly basis -- each person is checked for eligibility in each month of the simulation year. A person might be found eligible for Medicaid/CHIP in some months of the year but not the entire year. The eligibility rules are simulated in great detail, including the variations in eligibility rules across states. Like all TRIM3 modules, the MedicaidCHIP module can also simulate hypothetical Medicaid/CHIP rules.

There are some features of the Medicaid/CHIP programs that TRIM3 does not model. TRIM3 does not simulate eligibility for the institutionalized since they are not in the CPS universe. Further, children under age 15 generally cannot be simulated as eligible by disability, since they do not report their income and labor force information necessary to identify disability. However, the SSI module uses a special imputation process to identify certain children under 15 as receiving SSI. Only these children can be simulated by the MedicaidCHIP module as eligible by disability. The module does not simulate transitional benefits for those families no longer receiving cash Temporary Assistance to Needy Families (TANF) benefits due to increased income, increased employment, time limits, or other reasons, nor does it simulate any of the special pathways for working persons with disabilities. Finally, the module does not simulate eligibility for partial benefits.

This document describes the operation of the MedicaidCHIP module in detail. Note that the discussion refers to the many different "program rules" that control the operation of each TRIM3 MedicaidCHIP simulation. Details on each program rule and its potential values can be obtained from the TRIM3 Data Dictionary. The discussion is organized as follows:

Eligibility

Overview

There are many pathways through which a person can become eligible for Medicaid/CHIP. Whether a person is eligible is based on the category of person under consideration as well as their income and assets. For eligibility purposes, persons fall into one or more of the following categories:

  • Infant
  • Child
  • Parent/Caretaker
  • Pregnant woman
  • Person with disability
  • Elderly
  • Other adult
Eligibility criteria for each of these categories can vary by state. Starting in 2014, eligibility determination was simplified and made more standard across states for all but the elderly and disabled categories. This was accomplished by the ACA s requirement that a single definition of income (MAGI) be used, and that assets no longer be considered, when determining eligibility for pathways that are not targeted to the elderly/disabled (the Medically Needy pathway is also excluded from this requirement). Note that disabled persons can be tested for eligibility via a MAGI pathway if they are also in one of the MAGI categories. For pathways targeted towards elderly/disabled (as well as the Medically Needy pathway), income and assets are both taken into account.

Note that in TRIM3's terminology, an "eligible" person is someone who has passed all eligibility tests in a particular path, regardless of whether he/she actually enrolls to receive those benefits. In some administrative data, "eligible" means someone who not only passes eligibility tests, but took the further step of obtaining a Medicaid card (although they may not have actually used the card to obtain benefits). In TRIM3's terminology, this is a person who is "enrolled".

The eligibility pathways are described below in the order in which they are applied. In each month, the first eligibility path that makes a person eligible becomes that person's eligibility type for that month, and no further eligibility testing is performed for that month. A person's monthly eligibility type is stored in the result variable EligibilityType. It is possible for a person to be eligible through different pathways in different months of the year, and different members of the same family may be eligible under different pathways.

Non-Citizen Eligibility

Before a non-citizen can be considered for any of the Medicaid/CHIP eligibility pathways, he/she must meet certain non-citizen eligibility requirements. The MedicaidCHIP module follows the same procedure for determining non-citizen eligibility as other TRIM3 simulation modules (click here for details). The only modifications the MedicaidCHIP module makes to this standard method is that the national-level rule TempAlienEligible is replaced by the state-level rule StateTempAlienEligible, and legal aliens are always eligible for Medicaid/CHIP if they have been in the US for at least the number of years specified in the state-level rule EligIfInUSNumberOfYears (and if they pass the sponsor deeming test).

In addition to the standard method of determining the eligibility of non-citizens, the CHIPRA act of 2009 (section 214) gave states the option to extend coverage to lawfully present immigrant children and/or pregnant women regardless of how long they have been in the country, and regardless of whether they are subject to sponsor deeming. The state-level rule CHIPRA214 specifies whether a state applies this extension and whether it applies to children, pregnant women, or both, as well as whether it applies to both Medicaid and CHIP coverage or only to Medicaid coverage.

Eligibility Pathways for the Elderly and Persons with Disabilities (eligibility types 21-25)

The MedicaidCHIP module first checks if a person is eligible via one of the pathways available to elderly persons or persons with disabilities. The program rules that control the processing of these pathways are in the rule category Elderly & Disabled Eligibility and are divided into two sets of subcategories --- those applying to SSI-based eligibility pathways and those applying to poverty-based eligibility pathways. The SSI-based pathways are checked first. The eligibility rules controlling these pathways are in the subcategorySSI-based pathways while the rules specifying which microdata variables to use are in the subcategory SSI_based input variables. The various SSI-based pathways are checked in the following order:

SSI Cash Eligibility (eligibility type 21)

Receipt of federally funded SSI benefits (as opposed to supplemental state benefits) automatically makes an individual eligible for Medicaid. Information about SSI benefits received is passed from the SSI simulation via the program rules SSIBenefitsReceived and SSIFederalBenefitsEligFor. Any person simulated to receive benefits (SSIBenefitsReceived > 0) and some or all of those benefits were federal benefits (SSIFederalBenefitsEligFor > 0) is considered eligible for Medicaid.

Note that the program rule SSICashOption (in the group Mandatory Eligibility Pathways),gives the user the option of disabling this eligibility path (i.e. persons receiving federal SSI are not automatically made eligible for Medicaid).

Also note that for a few states, the asset and income limits for this path are actually more strict than those applied for federal SSI eligibility -- states that have expanded their SSI eligibility rules since 1972 are permitted to deny Medicaid eligibility to SSI recipients who would not have been eligible under their state's more restrictive 1972 rules. This is referred to as "Rule 209B", and is the only case in which a federally funded SSI recipient can be denied Medicaid eligibility. The rules in the group Rule 209b are used to indicate which states have more restrictve income and/or asset 209b rules in effect, as well as the levels of those more restrictive limits. The measures of income and assets calculated by the SSI module are used when determining if the 209b requirements are met. These amounts are passed from the SSI simulation via the rules SSISimulatedAvailableIncome and SSISimulatedUnitAssets.

SSI Cash Supplements Eligibility (eligibility type 22)

As described above, persons who are receiving federal SSI benefits must be covered by a state's Medicaid program. States may extend Medicaid eligibility to persons who, while not receiving federal SSI benefits, are receiving state supplements. The program rule DoesSSIStateSupQualifyForMcaid (in the group Optional Eligibility Pathways) indicates for each state whether or not it offers this extension. Information about a person's receipt of state supplements is obtained from the SSI simulation via the two program rules SSIBenefitsReceived and SSIFederalBenefitsEligFor (a person for whom SSIBenefitsReceived > 0 but SSIFederalBenefitsEligFor = 0 is receiving only state supplements). As is the case with mandatory SSI eligibility, eligibility is denied to persons who fail to meet their state's 209b restrictions (if any).

SSI non-cash Eligibility (eligibility type 23)

States may also extend Medicaid eligibility to persons who, while not receiving any SSI benefits (neither federal nor state), are eligible for either federal benefits or state supplements. The program rule DoesSSIEligQualifyForMedicaid (also in the group Optional Eligibility Pathways) indicates for each state whether or not it offers this extension. Information about a person's eligibility for federal or state SSI is passed from the SSI simulation via the two program rules SSIBenefitsEligibleFor and SSIFederalBenefitsEligFor. As is the case with other SSI-related eligibility pathways, eligibility is denied to persons who fail to meet their state's more restrictive 209b rules (if any).

Percent of Poverty Eligibility (eligibility types 24 & 25)

States have the option to extend eligibility to persons not eligible through any of the SSI-based pathways but with income and assets below a state-specified level. The rule group Poverty-based pathways specifies which states have adopted this option, as well as the state's income and asset limits. Note that income limits are specified as a percent-of-poverty (which can not exceed 100% of poverty) and can be different for elderly vs. disabled (whereas the asset limit is the same for both groups). The person's income and assets are generally measured the same as measured for determining SSI eligibility. However, differences in how non-citizens are treated in Medicaid vs SSI can cause these measures to differ. Consequently, the microdata variables used as input for the poverty-based pathways (specified by the rules in the subcategory Poverty-based input variables) must come from a different SSI simulation than the variables used for the SSI-based pathways. While in most respects identical to the simulation used for the SSI-based variables, the simulation used for the poverty-based pathways should use non-citizen eligibility rules that match the non-citizen rules being used in the Medicaid simulation.

MAGI-Based Eligibility Pathways (eligibility types 1-13)

The MedicaidCHIP module next checks if a person is eligible via one of the MAGI-based pathways. The program rules that control the processing of these pathways are in the rule category MAGI-based Eligibility. The MAGI-based pathways were created by the ACA, and apply to persons under age 65. All of these pathways apply only an income test (i.e. there is no limit on assets), and the income used is a monthly measure that follows the definition of "Modified Adjusted Gross Income" or MAGI.

The rules MonthlyMAGIComponents and MonthlyMAGIDeductions (in the subcategory Income Determination) specify what types of income are included in MAGI as well as which income types are deducted to arrive at the final measure of MAGI. In most cases, MAGI is calculated based on the composition of the tax unit (as defined in the FederalTax simulation module), although there are cases where an alternative unit definiton is used . Regardless of whether the tax unit or an alternative unit definition is used, the MAGI is converted to a percent-of-poverty using the poverty guideline for the unit's size, and then compared to a state-specific threshold to determine eligibility. Note that when converting MAGI to a percent-of-poverty, states must disregard from MAGI an amount equal to 5% of the poverty guideline. Since applying this disregard is the same as not applying it and instead increasing the threshold by 5 percentage points, a common practice when specifying MAGI thresholds is to list a threshold that is 5 percentage points above the state's "official" threshold. This practice is followed in the MedicaidCHIP module.

Thresholds vary by the type of person being considered, and are specified by the rules in the subcategory Thresholds. Note the following regarding these rules:

  • As mentioned, the thresholds are all 5 percentage points above the "official" thresholds.
  • One of the rules in the group Thresholds for Adults is the rule ExpansionStatus. This is not a threshold but instead shows which states have adopted the optional expansion of coverage to "other adults". However, note that even states that have not adopted the expansion may still have a non-zero threshold specified in the rule MAGIThresh_OtherAdults (although it can be less than the mandatory minimum of 138% required for states that have adopted the expansion).
  • Pregnant women not meeting the regular Medicaid MAGI threshold (as specified the rule MAGIThresh_Pregnant) may still be eligible if the state has a separate CHIP program that covers pregnant women (specified via the rule MAGIThresh_Pregnant_CHIP).
  • The thresholds in the Thresholds for Children group correspond to specific age ranges. These ranges are specified by the rules in the group Age Categories for Children (although the range applicable to the rule MAGIThresh_AdditionalChildren is fixed at 19-20).
  • If a state has a CHIP program for children, it can implement it either as a separate CHIP program, as a Medicaid expansion CHIP program, or as a combination of both. The rules MAGIThresh_Infants_CHIP, MAGIThresh_YoungerChildren_CHIP and MAGIThresh_OldererChildren_CHIP indicate if the state has implemented a Mediaid expansion CHIP program, while the rule MAGIThresh_Separate_CHIP indicates if the state has implemented a separate CHIP program.
When determining eligibility for CHIP (as opposed to regular Medicaid), the MedicaidCHIP module applies an additional restriction -- the person can not have other health insurance coverage that month. A person is considered to have "other coverage" in a month where he/she has:
  • Employer sponsored insurance. This information is input to the MedicaidCHIP module via the program rule MonthlyESICoverage, which usually specifies the EmployerSponsoredHealthIns module's monthly result variable CoveredByEmpHlthIns.
  • Medicare. This information is input to the MedicaidCHIP module via the program rule MedicareEnrollment, which usually specifies the Medicare module's annual result variable AnnualEnrollmentType. The module assumes that a person covered by Medicare is covered every month of the year.
  • CHAMPUS, VA or military health care. This information is input to the MedicaidCHIP module via direct reference to the input variable HealthChampusCoverage. The module assumes that a person with this type of coverage is covered every month of the year.
Note that this "other coverage" restriction for CHIP eligbility can be turned off via the rule CHIPOption.

If a person's MAGI is above all of the applicable MAGI-based thresholds, but their annual MAGI (i.e. the MAGI computed when filing their tax return, as opposed to the monthly MAGI computed for Medicaid/CHIP eligibility) is below 100% of poverty, the "safe harbor" rule is triggered. Under this rule, such a person can re-determine their Medicaid/CHIP MAGI-based eligibility by comparing their annual MAGI to the MAGI-based thresholds. The rule MAGIThresh_SafeHarbor2 specifies which states the "safe harbor" rule applies to, and gives the user the option to change the percent-of-poverty that triggers this rule in each state. Note that, like all MAGI-based eligibility pathways, the safe harbor rule does not apply to persons 65 or older. In addition, persons covered by Medicare are not covered by the "safe harbor" rule.

Medically Needy Eligibility (eligibility types 31-33 & 36-37)

States with Medically Needy programs cover persons whose income after medical expenses is under a state-specified threshold. This may include persons whose income is above the threshold prior to medical costs, but who "spend down" to below the threshold, as well as persons with very low medical expenses or no medical expenses whose income is low enough to fall below the threshold. Assets must also be below a state-specified level (but no adjustment is made to a person's assets to account for medical expenses). The program rules that control the processing of these pathways are in the rule category Medically Needy Eligibility.

Categorical Eligibility

States may choose whether or not to offer medically needy eligibility. If they do, they must provide coverage to pregnant women and children. States may choose to cover additional children up to age 21, parents and caretakers, the elderly, and disabled individuals. The program rule MedicallyNeedyGroups specifies for each state which of the following groups are categorically eligible:
  • Pregnant women and infants
  • Children up to a specified age (max 21)
  • Parents and Caretakers
  • Elderly according to SSI rules
  • Disabled according to SSI rules

Income & Asset Tests

Many of the states with a Medically Needy program have one set of income and asset limts for the elderly and disabled and another sets for persons in families (i.e. pregnant women, children, and parents/caretakers). Consequently, there are two groups of rules containing these limits --- the rule group Elderly & Disabled Limits and the rule group Family Limits (note that the rule group Elderly & Disabled Limits did not exist prior to version 17 of the MedicaidCHIP module, so the rule gropup Family Limits was used for all persons). When calculating income and assets for elderly/disabled persons, income and assets are generally measured the same as measured for determining SSI eligibility. However, as was described in the "Poverty-based eligibility" part of the "Eligibility Pathways for the Elderly and Persons with Disabilities" section, differences in how non-citizens are treated in Medicaid vs SSI can cause these measures to differ. Consequently, information about income and assets for elderly/disabled persons are obtained from the same set of variables used for the elderly/disable poverty-related pathways (i.e. the rules in the subcategory Poverty-based input variables). Also, note that the elderly/disabled limits only have values for units of size 1 or 2, since an SSI unit consists of either a single individual or a married couple. For persons in families, income is calculated by summing the variables specified by the rules EarnedIncome, ChildSupportIncome, and UnearnedIncome for all persons in the family (related subfamilies are treated as separate families). The result of this computation is stored in the monthly result variable FamilyIncome. Assets are calculated by summing the variables specified by the rules AssetIncome or AssetIncomeReporters for all members of the family. The result of this computation is stored in the monthly result variable MonthlyFamilyAssetIncome. An assumed rate of return of 6% is then applied to this amount to come up with an implied value of the family's assets.

Spend Down

The income amount is reduced by an amount meant to approximate the unit's medical expenses. The medical expense amounts used to simulate this "spend down" are specified by the rules in the group Average Medicaid Expenses, and vary by state and user group (children, adults, disabled, and elderly). Alternatively, if a microdata variable is available containing each person's medical expenses (specified via the program rule SpendDownAmounts), that information (plus an estimate of the Medicare Part B premiums paid by each individual) is used instead of the average Medicaid expenses. However, regardless of the expenses computed, in a baseline simulation, if an individual who falls within one of the groups covered by his/her state's medically needy program reported Medicaid coverage on the CPS but is not simulated as eligible that month through any of the non-medically needy rules, the MedicaidCHIP module assumes that person has medical expenses high enough to "spend-down" to within the medically needy income limits. Such persons are automatically considered to be income-eligible for the medically needy program, although they must still pass the asset test to be considered fully eligible. Such a person is also considered to be income-eligible for medically-needy coverage due to high medical expenses in all alternative simulations (assuming that the state is still being modeled as having a medically-needy program in the alternative). Note that the rule SpendDownOption can be used to turn off the application of spend down.

Eligibility of Unrelated Children (eligibility type 80)

Some children on the CPS are not related to any adults in the household. Since the income and asset level of children is generally based on the income received by their parents/guardians, the MedicaidCHIP module is unable to determine if these unrelated children are income/asset eligible for Medicaid/CHIP. Consequently, it is left up to the user to decide how TRIM should handle these children. If the rule UnrelatedChildOption is turned on (i.e. set to "1"), then these children are automatically considered to be eligible without any eligibility tests being performed. If this rule is turned off, all unrelated children are entirely excluded from Medicaid/CHIP eligibility. Note that UnrelatedChildOption only applies to unrelated children under age 15. Older unrelated children are treated the same as single adults. Also note that some of these unrelated children may be foster children, and foster children are almost always eligible for Medicaid. Children in this situation who are not formally foster children would likely be considered one-person units for Medicaid/CHIP eligibility.

Medicare Savings Programs (MSPs)

Low income persons covered by Medicare (i.e. low income elderly/disabled) may be eligible to have Medicaid pay some or all of their Medicare-related costs (i.e. premiums and cost-sharing). There are three types of MSP s that the MedicaidCHIP module models QMB, SLMB, and QI. A fourth type QDWI is not modeled. Furthermore, the MedicaidCHIP module excludes from MSP eligibility any Medicare enrollee who is simulated to be eligible for regular Medicaid (i.e. eligibility for the special categories of "QMB Plus" and "SLMB Plus" is not simulated).

There is a single asset limit for all three types of MSP s. The federal government sets a minimum level, but states can choose to have a higher limit or no limit. Prior to the passage of MIPPA (the Medicare Improvements for Patients and Providers Act of 2008), the minimum asset level had been a constant $4000 for single individuals and $6000 for couples. However, MIPPA changed that so that starting in 2010 the asset limit was the same as that for Medicare s Part D Low-Income Subsidy program (LIS) -- $6600 for singles and $9910 for couples. This asset limit is increased each year by the CPI (in years when the CPI decreases, the asset limit is not decreased but stays constant). If a Medicare recipient meets the asset requirement, the particular type of MSP he/she is eligible for is based on income as a percent of the poverty guideline. Again, there are federally-set minimum levels but states can implement higher levels. The federal levels are:

  • QMB 100%
  • SLMB 120%
  • QI 135%
Note that MSP eligibility does not use the MAGI definition of income. Instead, like the elderly/disabled poverty-related and Medically Needy pathways, income and assets are generally measured the same as measured when determining SSI eligibility. However, as described previously, differences in how non-citizens are treated in Medicaid vs SSI can cause these measures to differ. Consequently, information about income and assets for MSP eligibility is obtained from the same set of variables used for the elderly/disabled poverty-related and Medically Needy pathways (i.e. the rules in the category Elderly & Disabled Eligibility, subcategory Poverty-based input variables).

Among the various MSP programs, the QMB program is the most generous, and covers Medicare premiums as well as all cost-sharing. The SLMB program and the QI program both just cover premiums. The difference between the SLMB and QI programs (besides the higher income threshold for QI) is that the QI program is capped and is entirely financed with federal funds.

The rules in the category Medicare Savings Programs contain the asset limits and percent-of-poverty thresholds. The rule MSPAssets specifies each state's asset limit for MSP eligibility (999999 means the state does not impose an asset limit). The percent-of-poverty thresholds are specified via the three rules QMBPctOfPov, SLMBPctOfPov and QIPctOfPov. Note that D.C. only has a threshold specified for one of the three types of MSP's -- QMB. This is because D.C.'s threshold for QMB is generous enough to include persons who would normally only qualify only for SLMB or QI. Note that the results of MSP eligibility testing are stored in result variables separate from the variables used to store the results of Medicaid/CHIP eligibility testing (MSPAnnualEligibility and MSPMonthlyEligibility vs. AnnualEligibilityType and EligibilityType).

Enrollment

Overview

For each month that a person is simulated as eligible for Medicaid/CHIP, a decision can be made as to whether the person is actually enrolled in the program in that month. A person may be simulated to enroll in some but not all of his/her months of eligibility. If a person is simulated to enroll in a given month, he/she is assigned a value to the result variable EnrollmentType equal to the value of the result variable EligibilityType for that month. If the program rule EnrollmentOption is set to zero ("No special enrollment method used") the method for determining which months a person enrolls is based upon whether the MedicaidCHIP module is being run in baseline mode or in alternative mode, and is described below. The other possible values for the rule EnrollmentOption cause either all persons to enroll in all eligible months (option #1) or in no months (option #2).

Enrollment in a Baseline Simulation

During a baseline run, the determination of whether a person enrolls in an eligible month is made outside of the medicaid module. This requires that a SAS extract file be created and a series of SAS programs be run, the end result being the creation of a file indicating whether each person enrolls in each month. This file must be imported into TRIM as an "Out of Model" results file for the year being modelled. Once imported, the rule BaselineEnrollmentDecision should be set to point to the variable Enroll from this results file. When the MedicaidCHIP module needs to determine if a person is enrolled in a given month during a baseline simulation, it refers to this variable. Click here for details about the baseline enrollment processes.

Enrollment in an Alternative Simulation

Unlike the baseline enrollment decision, the enrollment decision for an alternative run is made within the MedicaidCHIP module. Consequently, alternative simulations can be run without needing to re-do the baseline enrollment procedure. In an alternative simulation, the monthly enrollment decision for persons who were eligible in that month in the baseline is the same as the baseline decision. Information on the baseline decision is obtained from the same "Out of Model" results file that was used in the baseline run, via the rule BaselineEnrollmentDecision. For persons who were not eligible in that month in the baseline, a logit equation is used to assign an enrollment probability. However, since the enrollment decision for newly-eligible persons is made jointly for all newly-eligible members of a family (subfamilies separate), these person-level probabilities must be combined into a single family-level probability. The method used to combine them is specified by the rule NewlyEligEnrProb:

  1. Use the maximum person-level probability of enrollment
  2. Use the minimum person-level probablity of enrollment
  3. Use the average person-level probability of enrollment

The resulting family-level probability is compared to a random number to determine if all the newly-eligible members of the family enroll in that month. The coefficients of the logit equation which is used to assign the person-level probabilties of enrollment to newly-eligible persons during an alternative run are as follows:

Dummy VariableCoefficient
Medicaid child0.106965
CHIP child-0.4787
Child age 0-0.55867
Child age 1-51.146442
Child age 6-121.052676
Child with family income <100% pov.1.907145
Child with family income 100-200% pov.0.492884
Adult Medicaid1.063281
Adult CHIP0.317048
Disabled Medicaid1.377276
Disabled, medically needy-0.17506
Disabled, age 19-440.095804
Elderly Medicaid-0.34152
Elderly, with family income <100% pov.0.31277

The resulting probabilties of enrollment are as follows:

Medicaid Childrenunder 100% pov100 - 200% pov200%+ pov
Child age 081%51%39%
Child age 1-596%85%78%
Child age 6-1296%84%76%
Child age 13+88%65%53%

CHIP Childrenunder100% pov100 - 200% pov200%+ pov
Child age 070%37%26%
Child age 1-593%76%66%
Child age 6-1292%74%64%
Child age 13+81%50%38%

Adults
Adult Medicaid74%
Adult CHIP58%

Disabled MedicaidMedically needyNot Medically Needy
Disabled age 19-4479%81%
Disabled age 45+77%80%

Elderly Medicaid
<100% poverty79%
100%+ poverty74%

Continuous Enrollment

In both baseline and alternative runs, after the initial monthly enrollment decisions have been made (as described above), the MedicaidCHIP module simulates the effect of "continuous enrollment" (also referred to as "continuous eligibility"). This allows a person to remain enrolled in Medicaid/CHIP for a certain number of months without requiring the person to report changes in income or other circumstances. Thus, under continuous enrollment rules, a person might be covered by Medicaid or CHIP in a month when s/he is technically ineligible based on their income in that month.

The program rules in the group ReportingPeriodGroup indicate the number of months of continuous enrollment offered by a state to persons in different categories:

  • RepPeriodKidsMcdExceptMedNeedy gives the months of continuous coverage for children eligible for Medicaid, with the exception of those eligible as Medically Needy
  • RepPeriodKidsCHIPMcdExtension gives the months of continuous coverage for children eligible under CHIP-funded Medicaid expansions
  • RepPeriodKidsCHIPSeparate gives the months of continuous coverage for children under separate-state CHIP programs
  • RepPeriodParents gives the months of continuous coverage for non-elderly adults who are parents
  • RepPeriodOtherAdults gives the months of continuous coverage for non-elderly adults who are not parents
  • RepPeriodKidsOption can hold state-specific options that cannot be captured by the standard rules

If the relevant reporting period rule is set to 1 month, there is no continuous enrollment. However, if it is set to a number greater than 1, continuous enrollment is simulated. In that case, once a person in that state/category is initially simulated to enroll in one month, his/her enrollment is automatically extended to the full reporting period. For example, if the rules indicate a 6 month reporting period for non-medically-needy Medicaid children, a child first simulated to enroll in March will be automatically simulated as enrolled through August. The reporting period associated with a person (regardless of whether they actually enroll) is stored in the monthly result variable ReportingPeriod. Note that a peson may not be able to "use" all of his/her months of continuous enrollment during the simulation year if the simulation year ends before the months are used. Since we do not model the opposite case when a person is eligible in January due to a spell of continuous enrollment that began in the prior year, the effects of continuous enrollment are slightly under-stated.

Note that pregnant women are always continuously enrolled for all months of their pregnancy.

The simulation of continuous enrollment is not affected by the simulated monthly eligibility status during the months of continuous enrollment. Once a person covered by continuous enrollment is simulated to enroll in a particular month, s/he is simulated to enroll for the rest of the continuous enrollment period, even if s/he is technically ineligible in some or all of the remaining months in the period. In addition, if the rule ContinuousEnrEligOpt is set to 0, the simulation of continuous enrollment does not affect the simulated eligibility status of the continuous enrollment months. Thus, the output variables might indicate that a person is both ineligible and enrolled in a particular month. Or, a person might be simulated as enrolled in Medicaid in a month when s/he is coded as eligible for CHIP, but not Medicaid. All such discrepancies in the output variables are due to continuous enrollment. However, if the rule ContinuousEnrEligOpt is set to 1, the eligibility type for the months of continuous enrollment is set to the enrollment type of the continuous enrollment spell. For example, in the above example, if when the child enrolled in March he/she was eligible for regular Medicaid (e.g. eligibility type #3 "MAGI Thresh children 1 to 5") but in the remaining 5 months of the continuous enrollment spell the child had eligibility type #7 "MAGI Thresh Separate CHIP" (or was not eligible at all), the eligibility type for those months would be set to #3. Note that the monthly result variable ContinuousEnrEligEffect is set to indicate what, if any, changes were made to each month's eligibility type due to continuous enrollment (see the TRIM3 Dictionary for details). Also note that the annual result variable NumberOfMonthsOfEligibility does not include months of continuous coverage where the person is technically ineligible, whereas the annual result variables NumMonthsNonSCHIPMcdElig and NumMonthsSCHIPElig do include such months.

Note that continuous enrollment is only simulated if enrollment is being simulated (i.e. if the simulation is set to just simulated eligibility, continuous enrollment is not simulated). Also, the current simulation of continuous enrollment does not capture some nuances in the continuous enrollment rules. In particular, we do not capture rules specifying that continuous CHIP enrollment ends if the person becomes enrolled in ESI and/or becomes eligible for Medicaid.

Additional Information

Annual Eligibility and Enrollment

In addition to determining eligibility and enrollment on a month-by-month basis, persons are also assigned an annual eligibility type and an annual enrollment type using an "ever-on" concept:

  • Annual Eligibility: If an individual is eligible for at least one month, s/he is considered to be eligible on an annual basis as well. If the person is eligible in more than one month, and the eligibility type varies from month to month, the MedicaidCHIP module first checks to see if the person is receiving federal SSI benefits or TANF benefits in any of his/her eligible months. If so, the eligibility type of the first such month is selected as the annual type. Otherwise, the monthly types are grouped into the following hierarchy of categories, and the type in the highest level (i.e. lowest number) is assigned as the annual type (if a tie, the type from the earliest of the tied months is chosen):
    1. Regular Medicaid (i.e. non-CHIP and not Medically Needy)
    2. Waiver (i.e. eligible through a state waiver program, not currently simulated)
    3. CHIP
    4. Medically Needy
  • Annual Enrollment: If an eligible person is simulated to enroll in at least one month, s/he is considered to be enrolled on an annual basis as well. If the person is enrolled in more than one month, and the enrollment type varies from month to month, the same process used to select an annual eligibility type is used to select an annual enrollment type. However, note that since it is possible for a person to be enrolled in some but not all of their eligible months, the type chosen for the annual enrollment type may be different from the type chosen for the annual eligibility type.

Reporter Status

For various purposes, the MedicaidCHIP module needs to know whether a person should be considered a "reporter"--i.e. did they report being covered by Medicaid/CHIP during the CPS interview. Due to the likelihood that many persons may be confused as to whether they are covered by Medicaid as opposed to CHIP (or vise-versa), the MedicaidCHIP module does not separate reporters into CHIP-reporters and Medicaid-reporters, but treats them all as undifferentiated "reporters". Thus, a child eligible for either Medicaid or CHIP is treated as a "reporter" if s/he is reported to be covered by either Medicaid or CHIP. For all persons, the result variable ReporterStatus indicates their "reporter status" as follows:

  • 0 = The person did not report being covered by Medicaid or CHIP.
  • 1 = The person reported either Medicaid or CHIP coverage, but this report was imputed ("allocated") by Census (i.e. the person was not a "true" reporter).
  • 2 = The person "truly" reported either Medicaid or CHIP coverage.

The determination of a person's reporter status is made as follows:

  • If either of the input variables corresponding to the CPS's direct questions about Medicaid and CHIP coverage (HealthCoveredMedicaid and CHIPCoverage) indicate that the person reported that type of coverage, and this was a "true" report (i.e. the corresponding variable AllocFlagMedicaidCoverage or AllocFlagCHIP has a value of "0"), then he/she is assigned a status of 2.
  • Otherwise, if any of the input variables corresponding to the CPS's "catch-all" questions about Medicaid and CHIP coverage indicate that the person reported coverage, and this was a "true" report, then he/she is assigned a status of 2. The "catch-all" questions are asked towards the end of the section of questions dealing with health insurance, and ask whether the person had "any other" health insurance that was not already mentioned. The variables corresponding to the "catch-all" questions are HealthCoveredOtherType1 - HealthCoveredOtherType6 and OtherHealthInsuranceCoverage1-OtherHealthInsuranceCoverage6. For each of these two groups of 6 variables, there is a single variable that indicates if the report was a "true" report (HealthOtherPlanTypeImputed and AllocFlagOtherHICoverage).
  • If none of these variables indicate that the person is a "true" reporter, but at least one indicates that he/she had reporter status imputed to them by Census, then he/she is assigned a status of 1.
  • Otherwise, he/she is assigned a status of 0.

Note that if after looking at the above variables, a person is flagged as a "true" reporter, a check is made of the variable AllocFlag665. This variable indicates whether all responses for this person were imputed by Census. If so, the person's status is changed from a 2 to a 1.

For more details about using the various health insurance variables on the CPS, see the article Using Health Insurance Coverage Variables.

While the result variable ReporterStatus is set for all persons, and is an annual variable, the result variable IsReporterThisMonth is monthly and is set only for persons who are reporters (i.e. ReporterStatus is 1 or 2) and are simulated as eligible for Medicaid/CHIP in at least one month. It indicates to which months their reported coverage is assumed to apply. This is determined by assigning coverage to a number of eligible months equal to the value of the variable HealthMedicaidMonthsCovered. If there are more eligible months than indicated by HealthMedicaidMonthsCovered, the eligible months are ranked according to the following heirarchy:

  1. Months in which federal SSI benefits or TANF benefits are received.
  2. Months in which the person is eligible for regular Medicaid (i.e. non-CHIP and not Medically Needy).
  3. Months in which the person is eligible through a state waiver program (not currently simulated).
  4. Months in which the person is eligible for CHIP.
  5. Months in which the person is eligible via the Medically Needy pathway.

Months in one category are chosen to be reporter months before months in later categories, until a number of eligible months equal to the value of HealthMedicaidMonthsCovered have been chosen. Note that if HealthMedicaidMonthsCovered is greater than the number of eligible months, or if it equals 0, then all eligible months are considered to be reporter months.

Pregnancy-based Eligibility

Although pregnancy is not reported on CPS data, Medicaid eligibility based on pregnancy may be simulated by the MedicaidCHIP module, based on the value of the rule PregnancyOption:

  • 0: Pregnancy is not simulated
  • 1: Pregnancy is imputed within the medicaid module and assigned to fixed months
  • 2: Pregnancy information is read in via the rule PregnancyIndicator
  • 3: Pregnancy is imputed within the medicaid module and assigned to random months.

Options #1 and # 3 can only be used during a baseline run. Both of these options impute pregnancy to all female family heads and spouses who have an infant (i.e. 0-year old) in their family. This method of imputation is referred to as the "0-year old" method. Using this method, the woman is assumed to be pregnant for the 9 months of April through December (option #1) or for a spell of 9 months begining in a random month (option #3). Note that in option #3, if the random start month is later than April, the spell is "wrapped around" to include January and later months as needed. In both options #1 and #3, a woman of child-bearing age (13-44) who reports receiving Medicaid/CHIP, but is not simulated to be eligible by any of the non-pregnancy-based rules in any month, is assumed to be pregnant in the 6 months of July through December (option #1) or for a spell of 9 months begining in a random month (option #3), if doing so makes her eligible in at least one of those months. This method is referred to as the "ineligible reporter" method.

In an alternative run, pregnancy can only be simulated using option #2. When this option is chosen, the rule PregnancyIndicator should refer to the monthly result variable IsPregnant produced by the baseline run. Option #2 can also be used in a baseline run if the user wishes to use an outside source of pregnancy information.

Alternative Simulations

An alternative (or non-baseline) simulation is one that models hypothetical or proposed rules rather than the actual rules that were in effect in a particular year. To set up an alternative simulation, do the following:
  • Set the rule SimulationMode to zero ("Alternative simulation").
  • Specify the monthly result variable EligibilityType from the baseline MedicaidCHIP simulation in the rule BaselineEligibilityType.
  • Specify the annual result variable AnnualEligibilityType from the baseline Medicaid simulation in the rule AnnualBaselineEligibilityType.
  • When simulating pregnancy, PregnancyOption should be set to "2" ("Pregnancy is read in via the PregnancyIndicator rule") and PregnancyIndicator should refer to the result variable IsPregnant from the baseline simulation.

Input Needed From Other Simulations

The MedicaidCHIP module requires input from numerous other simulations:
  • AlienPrep: Supplies information needed to determine the eligibility of non-citizens.
  • FedTaxImp: Supplies the variable KeoghContribution, which is used by MedicaidCHIP when calculating MAGI (see the rule MonthlyMAGIDeductions). If the tax match has not been done, use the _FedTaxImp_Blank version of these results, which sets KeoghContribution to 0. Note that before the 2018 baseline, this also supplied the variable ImputedCapitalGains, which was also used when calculating MAGI (see rule MonthlyMAGIComponents), but that information is now supplied by the FederalTax simulation as the variable NetCapitalGainsScheduled.
  • FederalTax: In addition to supplying the annual value of MAGI (ModifiedAGIForACA), and NetCapitalGainsScheduled (used for calculating monthly MAGI), the FederalTax simulation supplies the information needed to identify the MAGI unit (both the standard MAGI unit and the special alternative unit sometimes needed for Medicaid eligibility determination). If there is not a baseline available, a preliminary run can be used instead.
  • ChildSuuport: Needed to get the variable MonthlyCSinHhsWithoutKids. Usually a prep run is used.
  • SSI: Three SSI simulations are needed by the MedicaidCHIP module:
    1. The SSI result variable which indicates whether a person should be considered elderly or disabled on an annual basis is read in from the baseline SSI simulation (see the rule AnnualAgedAndDisabledStatus). The baseline simulation is also needed to supply the variables needed for determining eligibility via the pathways which directly link Medicaid eligibility to SSI eligibility (see the rules in the category Elderly & Disabled Eligibility, subcategory SSI-based pathways). These variables are read in by the rules in the category Elderly & Disabled Eligibility, subcategory SSI-based input variables.
    2. An alternative version of the SSI baseline is needed to supply the variables needed to determine eligibility via the elderly/disabled pathways other than the SSI pathways (i.e. the poverty-based pathways and the Medically Needy pathways). These variables are read in by the rules in the category Elderly & Disabled Eligibility, subcategory Poverty-based input variables. The alternative SSI simulation is identical to the baseline SSI simulation but has its non-citizen rules set to duplicate the MedicaidCHIP module s. To do this, change the rules in the category Categorical Eligibility, subcategory Citizenship Tests, to the same values to be used in the MedicaidCHIP baseline. Note that the rule ExtendedRefugeeEligYears does not exist in the MedicaidCHIP module, so set it to 0. Also, the National rule TempAlienEligible does not exist in the MedicaidCHIP module, but instead there is a state level rule StateTempAlienEligible. Usually this rule has the same value for all states, which should be used for the SSI rule TempAlienEligible. If the values differ among states, consult a senior researcher as to which value to set the SSI rule. Finally, make sure the SSI simulation is using at least version 42_1 of the module.
    3. A third SSI simulation is needed to supply a variable for the rule PreReformSSIVar. Before the 2018 baseline, the result variable PreReformSSI from an SSI simulation used as input to the SSI baseline was used. Since then, the variable PreReformSSI_Custom produced by a CustomOutput simulation has been used. This was done since the work of using the SSI module to impute some people as pre-PRWORA SSI recipients was no longer considered worth the effort since PRWORA was implemented such a long time ago.
    Note: As of version 31, the variable list input rules ElderlyDisabled_PersonTypeAnnual and SSIPassAssetTest are no longer used, even though they show up as using SSI results when clicking on the "Change Inputs" link when setting up a simulation.
  • TANF: Supplies the variable ChildCS_Final, which provides the value of child support after any withholding applied to TANF recipients, and the variable BenefitsReceived, which is used to identify cash recipients (i.e. persons receiving federal SSI or TANF benefits). A baseline simulation should be used. Cash recipient is mainly used for enrollment, but also for selecting which month to use to determine the annual eligibility type, as well as deciding which months should be considered reporter months for true reporters.
  • UnemploymentComp: Supplies the variable MonthlyUnemploymentComp, which contains the value of unemployment compensation received, after correction for under-reporting. A baseline simulation should usually be used.
  • EmpSponsoredHealthIns: Supplies the variable CoveredByEmpHlthIns. Since usually there is not an ESI baseline, a prep version is used instead. Fortunately, a version of the prep run for a prior year can be used to create the version for the current year since no rule or input changes are necessary for purposes of computing CoveredByEmpHlthIn (but note that the summary tables from this prep run will indicate that no one received ESI this is probably due to the new HI vars on the 2019 CPS-ASEC, but CoveredByEmpHlthIns appears to be unaffected).
  • Medicare: Supplies the variable AnnualEnrollmentType. We do not create a baseline Medicare simulation, so a prep version is used instead. A version of the prep run for a prior year can be used to create the version for the current year, but while no rule changes are necessary, it does need to use baseline SSI and TANF results.