The Urban Institute TRIM3 Reference Contact Us
The Urban Institute TRIM3 Reference Contact Us |
TRIM3 simulates eligibility for Medicaid and SCHIP on a monthly basis. In other words, each person or family is checked for eligibility in each month of the simulation year, and a person or family might be found eligible for Medicaid or SCHIP 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. The module is able to simulate both the rules that were in effect prior to the passage of the Personal Responsibility and Work Opportunity Act of 1996 (PRWORA) and the rules in place after PRWORA, including the post-PRWORA "Section 1931" eligibility rules. Like all TRIM3 modules, the Medicaid and SCHIP module can also simulate hypothetical Medicaid and SCHIP rules.
There are some features of the Medicaid and SCHIP 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 Medicaid as eligible by disability. The module does not simulate transitional Medicaid 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.
This document describes the operation of the Medicaid and SCHIP module in detail. Note that the discussion refers to the many different "program rules" that control the operation of each TRIM3 Medicaid/SCHIP simulation. Details on each program rule and its potential values can be obtained from the tools in the TRIM3 Navigator.
The discussion is divided into the following sections:
HOW TRIM3 SIMULATES MEDICAID AND SCHIP ELIGIBILITYTo determine if a family or any of its members is eligible for Medicaid or SCHIP, the module first tests whether any of the family members are automatically ineligible due to immigrant status. Medicaid follows the same procedure for determining non-citizen eligibility as other TRIM3 simulation modules (click here for details). The only modification Medicaid makes to this method is that the national-level rule TempAlienEligible is replaced by the state-level rule StateTempAlienEligible.
The module then simulates four different groups of eligibility tests:Mandatory, Optional, SCHIP, and Medically Needy. When a person or family fails all the tests in a particular category, Medicaid applies the next category of tests. The module first checks for mandatory eligibility, then state optional, then SCHIP, and finally medically needy eligibility (although note that by default, optional 1931 eligibility for children is tested for after SCHIP -- see the section Treatment of Optional 1931 Eligibles as SCHIP Eligibles for details). A person’s monthly eligibility type is the first type under which that individual qualifies. Different members of the same family may be eligible under different rules. Likewise, it is possible to be eligible for different reasons in different months of the year.
Some current and prior Medicaid eligibility rules are based in part on eligibility for or receipt of cash benefits. Medicaid eligibility for aged or disabled persons is generally linked with their eligibility for the Supplemental Security Income (SSI) program. Before August 1996, Medicaid eligibility for families with children was linked to their eligibility for Aid to Families with Dependent Children (AFDC), and since that time there has been some linkage with the Temporary Assistance to Needy Families (TANF) program (to a much lesser degree). Thus, to simulate Medicaid eligibility, TRIM3 uses not only the information from the survey input data on a family's income and other characteristics, but may also use TRIM-simulated data on whether a person is eligible for or receives benefits from SSI, TANF, or (in simulations of rules for 1996 or prior) AFDC. (Note that TRIM3 simulates both the pre-PRWORA AFDC rules and the post-PRWORA TANF rules in a single AFDC/TANF module.)
The following sections describe various aspects of TRIM3's simulation of eligibility in detail.
The amount of federal SSI benefits a person receives (according to TRIM3's SSI simulation) is passed to the Medicaid module via the rules SSIBenefitsReceived and SSIFederalBenefitsEligibleFor. The former indicates if any benefits were received, the latter indicates how much of that was federal (as opposed to state supplementary SSI benefits).
When simulating post-PRWORA rules, the user can use the rule Mandatory1931Eligibility to refer to results from a special AFDC/TANF run which simulates the mandatory component of 1931 eligibility rules. If this is done, then persons who meet the mandatory 1931 eligibility rules as modelled in that special AFDC/TANF simulation will be considered as mandatory Medicaid eligibles. Note that in most cases, the Mandatory1931Eligibility rule should not be used when the rule AfdcAutoElig indicates that AFDC/TANF recipients (and near-recipients) are mandatory eligibles.
If not eligible based on receipt of AFDC/TANF or mandatory Section 1931 rules, children, infants, and pregnant women may still be eligible under mandatory (national-level) "percent of poverty" rules. Children below the age specified in PovBasedMandAgeForChildren1 that also pass the income and asset tests specified in PovBasedMandPctForChildren1 and PovBasedMandAssetForChildren1, or children below the age specified in PovBasedMandAgeForChildren2 that also pass the income and asset tests specified in PovBasedMandPctForChildren2 and PovBasedMandAssetForChildren2, are mandatory eligibles. For pregnant women and infants (children under one year) the income and assets tests are specified in PovBasedMandAssetForPregAndInf, PovBasedMandPctForPregnant, and PovBasedMandPctForInfants.
The measure of income and assets used for these percent-of-poverty tests varies depending upon whether PRWORA rules are being simulated. For pre-PRWORA simulations, the income and assets of the person as determined for AFDC/TANF eligibility are used. They are passed from the AFDC/TANF simulation via the rules AFDCNetIncomeOfUnit and AFDCAssetsOfUnit. For post-PRWORA simulations, Medicaid calculates its own measure of income and assets, as described in the section Independently-Calculated Income and Assets.
A special group of children are those who are unrelated to any adults in the household. Since the income level of children is generally based on the income received by their parents/guardians, TRIM is unable to determine if these "unrelated children" are income eligible for Medicaid or SCHIP. Consequently, it is left up to the user to decide how TRIM should handle these children. If the national-level rule UnrelatedChildOption is turned on (i.e. set to "1"), then these children are automatically considered to be mandatory eligibles. Otherwise, they are excluded from all Medicaid and SCHIP eligibility.
States that supplement the federal SSI benefit levels have the option to extend Medicaid eligibility to those persons receiving the SSI state supplements. In TRIM3, the rule DoesSSIStateSupQualifyForMcaid allows the user to specify which states extend Medicaid eligibility to persons receiving an SSI state supplement. Information on whether a person received state supplements is passed to the Medicaid module via the rules SSIBenefitsReceived and SSIFederalBenefitsEligibleFor. (If the former amount exceeds the latter, then some of the benefits received were state supplements.)
States also have the option (implemented in TRIM3 via the rule DoesSSIEligQualifyForMedicaid) to extend Medicaid eligibility to persons eligible for (but not receiving) SSI benefits (either federal or state supplements). The amount of SSI benefits a person is eligible for (whether or not they are simulated to actually receive the benefit) is passed to the Medicaid module via the rule SSIBenefitsEligibleFor. Note that, as is the case with mandatory eligibility for recipients of federal SSI benefits, optional eligibility is denied to these SSI-linked groups if they fail to meet their state's "209b" restrictions (if any).
States may also extend eligibility to elderly or disabled individuals who fail to qualify for SSI due to their income. In TRIM3, such persons will be eligible if their income is below the percent-of-poverty specified by the rules PovBasedStatePctForDisabled and PovBasedStatePctForElderly, as long as they pass the SSI asset test. The amount of a person's income is based on the measure used for determining SSI eligibility, and is passed to the Medicaid module via the rule SSISimulatedAvailableIncome, while information as to whether they passed the SSI asset test is passed via the rule SSIPassAssetTest.
Under AFDC (and continuing under TANF) states had the option of not providing benefits (or not providing full benefits) to families with two non-disabled parents in which the primary wage-earner was unemployed (also called AFDC-UP units). If the state does not choose to include these families in their AFDC/TANF program, they still have the option to extend Medicaid eligibility to these families, via the rule DoesUPEligInNonUPQualForMcaid. In TRIM3, families with an unemployed head in such a state are eligible if they pass their state's AFDC/TANF income and asset tests and fail to meet the state's definition of a single or incapacitated-parent (IP) unit. Information about whether the AFDC/TANF income and asset tests were passed is passed to the Medicaid module via the rules AFDCPassAssetTest, AFDCPassNetIncomeTest, and AFDCPassGrossIncomeTest. Information on whether the individual's family qualified as a single-headed, UP, or IP unit is passed to the Medicaid module via the rule AFDCPassParentTest.
Before PRWORA, states also had the option to extend Medicaid eligibility to individuals eligible for AFDC/TANF benefits but not receiving any. In TRIM3, the rule DoesAFDCEligQualifyForMedicaid is used (pre-PRWORA) to indicate which states extend Medicaid eligibility to eligible non-recipients. After the passage of PRWORA in 1996, the purpose of this rule shifted to simulating a simplified version of section 1931 eligibility rules. In order to do this, the definition of "eligible for AFDC" was extended so that in some cases optional Medicaid eligibility was granted even to persons not actually eligible for a benefit as long as they passed all the other AFDC/TANF requirements (i.e. after passing all other tests, the AFDC/TANF benefit-calculation formula returns a zero or negative benefit amount). The amount of AFDC/TANF benefit a person is simulated to be eligible for (whether or not they are simulated to actually receive the benefit) is passed to the Medicaid module via the rule AFDCBenefitsEligFor. In the case where passing the benefit test is not necessary, information on whether the other tests were passed is given to the Medicaid module via the rules AFDCPassParentTest, AFDCPersonType, AFDCPassAssetTest, AFDCPassNetIncomeTest, and AFDCPassGrossIncomeTest.
A special category of optional Medicaid eligibility for children is the "Ribicoff" category. Under this category, states can extend Medicaid eligibility to AFDC/TANF children (under a certain age) who pass the state's AFDC/TANF asset and income tests but not the categorical (or "parent") tests (i.e. the child is neither in a single-parent family nor in a two-parent family qualifying as IP or UP). The age cutoff for each state is specified by the rule RibicoffChildrenStateAge, and information on whether a child is simulated to be an AFDC/TANF child is given to the Medicaid module via the rule AFDCPersonType. Information on whether the asset, income, and parent tests were subsequently simulated to be passed for these children is given to the Medicaid module via the rules AFDCPassAssetTest, AFDCPassNetIncomeTest, AFDCPassGrossIncomeTest, and AFDCPassParentTest. Note that after the passage of PRWORA in 1996, the AFDC income and asset tests that the child must pass are based on 1996 AFDC levels, not the current year's levels, while the parent test remains linked to the current AFDC/TANF requirements.
States may extend "percent-of-poverty" eligibility by increasing the age, income, and/or asset limits beyond the minimum mandatory federal limits. In TRIM3, these higher limits can be specified through the rules PovBasedStatePctForChildren1, PovBasedStatePctForChildren2, PovBasedStatePctForPregnant, PovBasedStatePctForInfants, PovBasedStateAssetForChildren1, PovBasedStateAssetForChildren2, PovBasedStateAssetForPrgAndInf, and PovBasedStateAgeForChildren1, PovBasedStateAgeForChildren2. However, the same measure of income and assets is used as in the federally mandated tests.
States may extend their 1931 eligibility criteria beyond the mandatory minimum levels. The user can use the rule Optional1931Eligibility to refer to the results of a special AFDC/TANF run which simulates these expanded (or "optional") 1931 eligibility rules. If this is done, then persons who meet the optional 1931 eligibility rules as modelled in that special AFDC/TANF simulation will be considered as optional Medicaid eligibles. Note that in most cases, the Optional1931Eligibility rule should not be used when the rules DoesUPEligInNonUPQualForMcaid and DoesAFDCEligQualifyForMedicaid indicate that persons may be covered by these rules.
Section 1931 rules originated in the welfare reform provision of the PRWORA. This act eliminated the AFDC program and replaced it with the new TANF program. While ending the direct link between receipt of AFDC/TANF cash benefits and eligibility for Medicaid, this act also added a new Section 1931 to the Social Security Act. Under Section 1931 of the Act, states are required to provide Medicaid eligibility to low-income families who meet the pre-PRWORA AFDC income and resource standards and other requirements that were in effect on July 16, 1996. However, states can diverge from the AFDC plans in effect on July 16, 1996 as follows:
Beginning with version 21 (1999), TRIM3 simulates Section 1931 eligibility by performing special runs of the AFDC/TANF module that capture each state's Section 1931 eligibility requirements, including complex rules for income disregards (excluding certain types of income such as child care or work expenses when determining eligibility), asset testing (including the value of certain assets when determining eligibility), and categorical eligibility requirements. Although most states have deviated from the minimum criteria (1996 AFDC eligibility rules) for Section 1931, the basic structure of the eligibility tests remains the same, so the AFDC/TANF module is best suited to simulate 1931 rules.
As described above, the minimum Section 1931 eligibility criteria are modeled using a "Mandatory" Section 1931 simulation run of AFDC/TANF, which is essentially the same as the 1996 baseline AFDC run. The eligibility results of this run are passed to the Medicaid module via the rule Mandatory1931Eligible. Deviations from the minimum criteria are modeled using an "Optional" Section 1931 simulation run of AFDC/TANF, which incorporates any expansions and/or deviations from the 1996 AFDC eligibility rules. Results of this run are passed to Medicaid via the rule Optional1931Eligible. If there are variables specified by the Mandatory1931Eligible and Optional1931Eligible input variable rules, then those variables are used to indicate mandatory or optional Section 1931 eligibility. Otherwise, if no variables are specified, no Section 1931 eligibility is modeled.
Note that in TRIM3's summary tables, children eligible as optional 1931 eligibles are categorized as SCHIP eligible. See the section Treatment of Optional 1931 Eligibles as SCHIP Eligibles for details.
If CHIPOption is set to "2", SCHIP eligibility is simulated the same as above, with the exception that children who are covered by employer sponsored insurance (ESI), Medicare, Champus, or other military health care are ineligible for SCHIP coverage. Note that access to the Indian Health Service is not considered health insurance, and does not preclude SCHIP eligibility.
The CPS variables indicating ESI, Medicare, Champus, or military health coverage are all annual, not monthly. We assume that a child reported to be covered by Medicare, Champus, or other military health care is covered the entire year. However, that assumption would be less appropriate for ESI. Therefore, TRIM assumes that a child covered by ESI is covered only in those months when one or both of his/her parents/guardians are working.
TRIM3 simulates the two main types of SCHIP programs: programs that are actually extensions (or expansions) to Medicaid, and programs that are separate from Medicaid. A state may operate only an SCHIP-funded Medicaid expansion program, only a separate state SCHIP program, or both. Some states charge a premium for some or all SCHIP coverage. States may also have different SCHIP rules for children in different age ranges.
| Medicaid Extension Programs | Separate State Programs | |||
|---|---|---|---|---|
| Minimum Age | Maximum Age | Minimum Age | Maximum Age | |
| Age Range # 1 | McdExtLowAge1 | McdExtHighAge1 | CHIPLowAge1 | CHIPHighAge1 |
| Age Range # 2 | McdExtLowAge2 | McdExtHighAge2 | CHIPLowAge2 | CHIPHighAge2 |
| Medicaid Extension Programs | Separate State Programs | |||
|---|---|---|---|---|
| Free | Premium | Free | Premium | |
| Age Range # 1 | McdExtFreePct1 | McdExtPremPct1 | CHIPFreePct1 | CHIPPremPct1 |
| Age Range # 2 | McdExtFreePct2 | McdExtPremPct2 | CHIPFreePct2 | CHIPPremPct2 |
| Method for Determining Disregard/Deduction | Amount to Disregard/Deduct (if applicable) | |
|---|---|---|
| Child Care Expenses | CHIPChildCareDeductOption | CHIPMaxChildCareDeduct |
| Child Support Income | CHIPChildSuppDisregardOption | CHIPMaxChildSuppDisregard |
| Work Related Expenses | CHIPWorkExpDisregardOption | CHIPWorkExpDisregard |
The income amounts that these deductions and disregards are applied to are the same ones used in simulating eligibility for Medicaid. They are specified by the rules FamilyEarnedIncome, FamilyChildSupportIncome, and FamilyUnearnedIncome. Although similar, the above disregards and deductions used for SCHIP are separate from those used when Medicaid calculates income for non-SCHIP eligibility (see the section Independently-Calculated Income and Assets). The final value of family income for SCHIP eligibility is calculated as follows:
Family income =
Earned income less the work expense disregard +
Unearned income +
Child support income less the disregarded amount –
The amount of child care expenses disregarded
States choose whether or not to offer "medically needy" eligibility. If they do, they may provide that type of eligibility to any combination of the following types of people:
Income and asset tests must also be passed, as specified in the state-specific rules MedNeedyIncomeLimitSize01-MedNeedyIncomeLimitSize10 and MedNeedyAssetLimitSize1-MedNeedyAssetLimitSize4, depending upon the family size. The national-level rule IncomeCalculationType indicates whether income and assets are obtained from the SSI and AFDC/TANF modules, or if they are calculated using Medicaid's own methods, as described in the section Independently Calculated Income. If income and assets are being obtained from the SSI and AFDC/TANF modules, the income amounts are specified by the rules SSISimulatedAvailableIncome) and AFDCNetIncomeOfUnit). The income amount is then reduced by an amount meant to approximate medical expenses. The medical expense amounts used to simulate "spend down" are specified by the rules beginning "AvgMedicaidExpFor...", and vary by state and user group (children, adults, disabled, and elderly). The value of assets is obtained either from the Medicaid/SCHIP module's own calculations or from the SSI and AFDC/TANF modules (via the rules SSISimulatedUnitAssets) and AFDCAssetsOfUnit).
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 benefits on the CPS but is not simulated as eligible through any of the non-medically needy rules, TRIM assumes that person has medical expenses higher than the average and high enough to "spend-down" to within the limits. Such persons are automatically considered by TRIM3 to be medically needy, and the income and asset tests are not performed. Such a person is also considered to be 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).
In order to simulate these changes, some new rules were added to the Medicaid module, and the values of some old rules were changed. It is important to note that the incorporation of PRWORA rules into the Medicaid module was done in two phases. The first phase encompassed the 1997 and 1998 baselines. In these years states had not diverged enough from pre-PRWORA eligibility criteria to warrant anything more than a simplified approach. In this phase, Medicaid simulations kept the direct link between TANF receipt and Medicaid eligibility (i.e. the rule AfdcAutoElig=Yes), and (as described in the 1931 section), used the rule DoesAFDCEligQualifyForMedicaid to simulate the new 1931 eligibility criteria. The only other change was to de-link the calculation of income and assets from the TANF-calculated values when determining percent-of-poverty eligibility. By setting the rule IncomeCalculationType to 2, income and assets are calculated using Medicaid's own method, as described in the section Independently Calculated Income.
By 1999, this simplified approach was no longer sufficient. One enhancement (as described in the 1931 section) was to move the simulation of the 1931 rules to special runs of the AFDC/TANF module, where they can be modelled more accurately. The results of these runs are passed to Medicaid via the rules Mandatory1931Eligible and Optional1931Eligible. In order for this to work properly, the rules AfdcAutoElig, DoesUPEligInNonUPQualForMcaid, DoesAFDCEligQualifyForMedicaid, and BenefitsUnitEligForBelowMin must be turned off. The computation of income and assets for the percent-of-poverty rules continued to be done using Medicaid's own method, by setting IncomeCalcualtionType to 2. To simulate the fact that Ribicoff and Medically Needy eligibility are now linked to 1996 AFDC eligibility levels, the following rules in the "AFDC Info" category were changed to point to results from a 1996 AFDC/TANF run (in order to minimize the number of special AFDC/TANF runs required, these rules were set to point to the Mandatory 1931 run, which is essentially the same as a 1996 AFDC baseline):
For post-PRWORA runs, the rule IncomeCalculationType should be set to 1. This begins by computing the family's gross income as above but adding any income AFDC/TANF considered deemed to the family (AfdcDeemedIncome). This new total is then reduced by the deductions and disregards specified by several state-specific rules. A fixed amount is disregard from earnings (FamilyEarningsFixedDisregard), a fraction is disregard from the remainder (FamilyEarningsFractDisregard), a fixed amount per child is disregarded from child support income (FamilyChildSupportDisregard), and a fixed amount of child care expenses is disregarded from the result(FamilyChildCareExpDisregard) to produce a net family income as follows:
Family income =
Earned income less the disregarded amount +
Any
income deemed to the family according to AFDC rules +
Unearned income +
Child support income less the disregarded amount –
The amount of child
care expenses disregarded
Note that in the TRIM3 module, the disregard policies are applied to the entire year even if in the "real world" the disregards are only applied to part of the year (for example, the first four months). Also, note that even when IncomeCalculationType = 1, a simulation could be run using gross income by setting all disregards to zero.
One nuance in the definition of income is whether or not monthly fluctuations in income are "smoothed" in determining monthly eligibility. Through the state-specific program rule ApplyIncomeSmoothing, TRIM can simulate one type of "smoothing"-- smoothing out variations in earnings received in a particular month that are due solely to the receipt of weekly or biweekly (as opposed to monthly) paychecks. If the TRIM conversion procedures for a particular year treated some months as having 5 weeks and others as having 4 weeks, the monthly earnings amounts created by TRIM assumed weekly paychecks, meaning that for a full-year worker, earnings in 4-week months are only 80 percent of the amount assigned to a 5-week months.
When "turned on", earnings smoothing ensures that variations in earnings due only to the weeks in a month do not affect eligibility. For people working in all weeks of a month, earnings in four-week months are multiplied by an adjustment factor of 1.0833 and earnings in five-week months are multiplied by an adjustment factor of .8667. The sum of smoothed monthly earnings will be equal to actual, unsmoothed annual earnings for persons working in all 52 weeks of the year. However, the sum of smoothed monthly earnings for persons working less than 52 weeks may be slightly lower or higher than actual annual earnings, depending on how the weeks of employment are distributed over the year; note that the goal is to approximate the earnings that would be used by transfer programs for eligibility and benefit purposes, rather than to create variables to represent actual monthly earnings.
Different regression equations are used to estimate insurance values for different user groups and for different health expenditure categories. The user groups are: children (under 19, and not head/spouse of a family, and not pregnant, and not disabled), adults (not disabled, and 19-64 or under 19 but head/spouse of a family or pregnant), disabled (under 65 and disabled according to the SSI module’s definition), and elderly (anyone 65 or older). The expenditure categories are: mandatory Medicaid costs, drug costs, long-term care, home health care, and other costs. (Long-term care and home health care insurance values are estimated only for aged and disabled persons.) The explanatory variables include sex, race, age, number of months eligible for Medicaid during the year, whether living in a rural location, receipt of AFDC/TANF or SSI benefits, whether eligible for Medicaid under the Medically Needy program, whether the head of the family or a dependent (for the groups under age 21 and 21-64). The equations were estimated using actual Medicaid expenditures from HCFA’s "tape-to-tape" data system.
After the regression equations have been evaluated for a particular Medicaid eligible, adjustment factors are applied. The factors vary by user group, expenditure type and by state. (They are coded in the state-specific rules beginning AdjAdult..., AdjAged..., AdjChild..., and AdjDisabled.) The adjustment factors serve three purposes: simulating the state-level variation in insurance costs, adjusting for price increases since the year of the data used to estimate the equations, and aligning to targets. Targets are computed by user group, state, and expenditure category. (The targets are coded in the state-specific rules beginning TgtAdult..., TgtAged..., TgtChild..., and TgtDisabled.) Adjustment factors are established for each year during that year's baseline simulation.
Eligible persons who are reported to be enrolled in Medicaid or SCHIP according to the CPS interview ("eligible reporters") are almost always included in the simulated caseload for the number of months of reported enrollment. (The exception to this rule is discussed below.)
Eligible individuals who receive cash aid (SSI or AFDC/TANF) in a particular month are always enrolled in Medicaid or SCHIP in that month. This assumes that an individual who receives cash aid would be aware of and would not refuse medical assistance. Information on monthly benefits received is passed to the Medicaid module via the rules SSIBenefitsReceived and AFDCBenefitsReceived.
Additional eligible individuals are included in the simulated caseload probabilistically in order to reach administrative targets, which vary by user group (children, adults, disabled, and elderly) and by state. For each eligible person who is neither a reporter nor a cash recipient, the module obtains the appropriate probability of participation based on his/her user group and state of residence. (See the discussion of adjustment factors and administrative targets below for more details.) If the person's random number for purposes of Medicaid/SCHIP eligibility is less than the probability of participation, the person will be included in the simulated caseload.
Persons who report Medicaid coverage but who seem ineligible for the program in every month are never simulated to enroll. As discussed above, however, some reporters who initially appear ineligible are treated as eligible on the assumption that they are either pregnant or eligible as Medically Needy due to very high medical expenses.)
TRIM3 attempts to make the simulated enrollment decisions consistent among members of a family in a given month. All family members who are eligible for Medicaid and who report Medicaid are generally either enrolled or not enrolled as a group; the same is generally true for all family members who are eligible but who do not report Medicaid coverage. The way that the module accomplishes this is by using the same random number for purposes of Medicaid/SCHIP enrollment for all eligible persons in the same family. The random number that is used for all eligible people in the family is the random number for purposes of Medicaid/SCHIP eligibility that was assigned to the first eligible person in the family.
Three state-specific program rules indicate the number of months of continuous coverage offered by a state to children in different categories. RepPeriodKidsMcdExceptMedNeedy gives the months of continuous coverage for children eligible for Medicaid, with the exception of those eligible as medically-needy; RepPeriodKidsSCHIPMcdExtension gives the months of continuous coverage for children eligible under SCHIP-funded Medicaid expansions; and RepPeriodKidsSCHIPSeparate specifies continuous coverage of children under separate-state SCHIP programs. In addition, RepPeriodKidsOption can hold state-specific options that cannot be captured by the standard rules.
If the relevant continuous-coverage rule is set to 1 month, there is no continuous enrollment. However, if it is set to a number greater than 1, continuous coverage is simulated. In that case, once a child in that state/category is initially simulated to enroll in one month, his/her enrollment is automatically extended to the full continuous-coverage period. For example, if the rules indicate 6 months of continuous coverage for non-medically-needy Medicaid children, a child first simulated to enroll in March will be automatically simulated as enrolled through August. Note that a child may not be able to "use" all of his/her months of continuous coverage during the simulation year, if the simulation year ends before the months are used. Since we do not model the opposite case when a child is eligible in January due to a spell of continuous coverage that began in the prior year, the effects of continuous coverage are slightly under-stated.
Note that TRIM3's simulation of continuous enrollment is not affected by the simulated monthly eligibility status during the months of continuous enrollment. Once a child 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. Further, the simulation of continuous enrollment does not affect the simulated eligibility variables, as currently programmed. Thus, the output variables might indicate that a child is both ineligible and enrolled in a particular month. Or, a child might be simulated as enrolled in Medicaid in a month when s/he is coded as eligible for SCHIP, but not Medicaid. All such discrepancies in the output variables are due to continuous coverage.
The current simulation of continuous enrollment does not currently capture some nuances in the continuous coverage rules. In particular, we do not capture rules specifying that a child’s continuous SCHIP enrollment ends if s/he becomes enrolled in ESI and/or if s/he becomes eligible for Medicaid.
The adjustment factors are specified by the following state-specific rules:
Another set of state-specific rules specifies annual (or "ever-on" targets). (Note that there are no annual citizenship targets.)
An example of how the adjustment process functions: Imagine that for a particular type of user-group in a particular state, the target is 100 enrollees, and (in the month being processed) there are 60 AFDC/TANF and SSI recipients, 30 other eligible reporters, and 50 other persons who are eligible but who did not report Medicaid receipt. The enrollment adjustment factor would be set to .20. The model would assign as enrollees that month all 60 AFDC/TANF and SSI recipients, all 30 other eligible reporters, and 20% (10 out of 50) of the eligible non-reporters. Note that even though the enrollment decision is made monthly (and the mix of cash-recipients, eligible-reporters, and eligible non-reporters may vary from month-to-month), the adjustment factors do not vary by month (since we are trying to hit an average-monthly target rather than month-specific targets).
A baseline simulation can be performed in order to emphasize closeness to the average monthly targets or the ever-on targets. In either case, the simulation results can be tabulated both ways--showing both average monthly and ever-on simulated caseload. There is currently no method that allows us to reach both sets of targets simultaneously.
Note that this aspect of the Medicaid enrollment decision (not enrolling some eligible reporters in some cases) differs from the approaches used in other TRIM3 transfer program modules (which always allow all eligible reporters to participate even if targets are exceeded). Different choices have been made for different modules due to differences in the ways that module results have been used.
If Medicaid eligibility due to pregnancy is being modeled, TRIM3 initially treats as pregnant all female family heads/spouses who have an infant (i.e. 0-year old) in their family. In addition, in a baseline simulation, a woman is considered to be eligible due to pregnancy if she is of child-bearing age (i.e. 13-44), she reports receiving Medicaid, but she is not simulated to be eligible by any of the non-pregnancy rules. Note that a female Medicaid reporter of child-bearing age who is also a parent, and who is not simulated as eligible due to other rules, is assumed eligible as medically-needy (rather than eligible due to pregnancy) if she lives in a state with medically-needy coverage for caretaker relatives. Note also that in an alternative simulation, the identification of pregnancy is taken from the baseline simulation.
Once covered by Medicaid, a pregnant woman must be covered throughout her pregnancy. Therefore, if eligibility is simulated during one month of the pregnancy, the module assumes the woman is eligible in all subsequent months.
The input variable HealthMedicaidCoverage is created during the conversion of CPS data to TRIM3 format. It is set to "covered" when either the variable HealthCoveredMedicaid indicates coverage or if Medicaid (or, starting with the March 2000 CPS, SCHIP) is reported in the "catch-all" questions about additional health insurance coverage (HealthCoveredOtherType1-6). Also starting with the March 2000 CPS, a separate variable is available (CHIPCoverage) that indicates reported SCHIP coverage. Due to the likelihood that many persons may be confused as to whether they are covered by Medicaid as opposed to SCHIP (or vise-versa), TRIM3 does not separate reporters into SCHIP-reporters and Medicaid-reporters, but just treats them all as undifferentiated "reporters". Thus, a child eligible for either Medicaid or SCHIP is treated as a "reporter" if s/he is reported to be covered by either Medicaid or SCHIP.
When referenced by the Medicaid module, these variables (HealthMedicaidCoverage and CHIPCoverage) indicate whether Census considers the person to be covered. However, this doesn't necessarily mean that the person actually reported that they were covered, since Census sometimes imputes (or "allocates") a positive response to the coverage questions if the person either didn't answer them or the answer conflicts with other information.
For TRIM3's purposes, we don't want to consider such "allocated" persons as covered. Thus, the next step to determining reporter status is to take those persons who are indicated to be covered by HealthMedicaidCoverage or CHIPCoverage and drop out those that had coverage allocated to them. Allocated coverage is indicated via special variables called "allocation flags". The allocation flag AllocFlagMedicaidCoverage applies to the variable HealthCoveredMedicaid, the flag AllocFlagCHIP applies to the variable CHIPCoverage, and the allocation flag HealthOtherPlanTypeImputed applies to all the "catch-all" questions.
If both AllocFlagMedicaidCoverage and AllocFlagCHIP indicate that allocation was done, then the reporter is dropped. If not dropped, but HealthCoveredMedicaid indicates no coverage, then this means the coverage indicated by HealthMedicaidCoverage was obtained from the "catch-all" questions. In this case, the "catch-all" allocation flag HealthOtherPlanTypeImputed needs to be checked as well, and if it indicates that allocation was done, the reporter is dropped. An additional allocation flag is available -- AllocFlag665 -- which indicates whether the entire record (rather than just the response to a particular question) was allocated. If this flag indicates allocation, then the reporter is also dropped.
Special variables were available before the March 1995 CPS which indicated whether a child (i.e. person < 15) was covered by Medicaid. Consequently, when using 1993 or earlier input data, children are considered reporters if they either pass the above criteria or:
If HealthChildCoveredbyMedic = 1 and AllocFlagMedicChildren not = 1 and AllocFlagMedicNumberOfChildren not = 1.
The annual output variable IsReporter indicates whether a person was considered to be a Medicaid reporter by the TRIM3 simulation, and is calculated for all persons, regardless of their Medicaid eligibility. The monthly variable IsReporterThisMonth indicates (for reporters) which months TRIM3 imputes the reported coverage to apply to. This is done by assigning coverage to a number of the eligibles months equal to the variable HealthMedicaidMonthsCovered. If there are more eligible months than HealthMedicaidMonthsCovered, months are selected by ranking them according to their eligibility category as follows:
These results variables can exported from TRIM3 and used to obtain estimates of both annual and FYE (i.e. average-monthly) Medicaid eligibility. When estimating FYE numbers, the user can use either of two methods. One method is to simply adjust the weight of individuals simulated eligible (on an annual basis) by the number of months they were eligible. This information is contained in the result variable NumberOfMonthsOfEligibility. While this method is sufficient for FYE estimates of eligibility versus non-eligibility, it does not allow for estimates of eligibility by the type of eligibility. In order to get that information, the user must actually look at the month-by-month eligibility data (contained in the monthly variable EligibilityType). The user should be aware that this method, because it involves analysis of monthly data, is more resource-intensive than the first method.
When using simulated enrollment variables, the user should be aware that the Medicaid simulation which created the variables could have aligned simulated enrollment to either annual or FYE published totals. Generally, runs for years before 1997 were aligned to annual totals, while 1997 and after are aligned to FYE totals. Check with the user responsible for the run to find out for certain.