Skip to main content

Documenting Citizenship and Identity Using Data Matches

A Promising Strategy For State Medicaid Programs

Under the Deficit Reduction Act of 2005 (DRA), U.S. citizens must document their citizenship and identity when applying for Medicaid or renewing their coverage. The new requirement, which took effect on July 1, could jeopardize the health coverage of substantial numbers of eligible citizens who are unable to secure the required documentation. [1] The significant challenges that many citizens — primarily children, parents, pregnant women and people with disabilities — will face as they attempt to obtain the required documents could result in their health coverage being delayed or denied. That, in turn, could deprive them of preventive and prenatal care and other necessary medical services.

The requirement also creates new administrative burdens for states, since it applies to all current Medicaid beneficiaries and all applicants except those who participate in both Medicaid and Medicare (so-called “dual eligibles”) or are enrolled in Medicaid as a result of receiving Supplemental Security Income (SSI) benefits. Much of this burden will be felt over the course of the coming year, during which about 38 million U.S.-born adults and children currently enrolled in Medicaid will be subject to the new requirement when they seek to renew their coverage. In addition, about 10 million otherwise eligible U.S. citizens can be expected to apply for Medicaid each year. Applicants unable to produce the documents will be denied coverage.

Implementing the new requirement thus will pose daunting challenges for states. States can, however, substantially reduce the burdens on themselves, and on applicants and beneficiaries as well, by adopting an option made available in the interim final regulation that the Centers for Medicare and Medicaid Services (CMS) issued on July 12.[2] This option allows states to obtain evidence of a person’s U.S. citizenship and identity by conducting electronic cross-matches with existing databases, such as vital records, Social Security, the Food Stamp Program, and the state motor vehicles department.

By adopting this option, states can free many people from the need to obtain and submit the required documents. Adopting the option also will minimize the increased costs that state and local agencies will face in conducting eligibility determinations. For example, the option can help states retain mail-in enrollment and renewal systems and thereby lessen or avoid the need to hire additional workers to handle an influx of in-person applications. (If the option is not adopted and many applicants and beneficiaries are unwilling to mail original copies of important documents and risk having them lost, mail-in enrollment could deteriorate.) In addition, by enabling eligible U.S. citizens to retain or promptly gain Medicaid coverage, data matching can help to ensure that health care providers receive Medicaid payments for services rendered.

This paper describes how some states are implementing or planning to implement the electronic data match option. It also explains the technological and logistical challenges that some states may encounter as they explore the possibility of conducting such matches and some possible solutions. Not all states have the capacity to fully avail themselves of this option, however.

Cross-Matching with Vital Records Could Document Most Individuals’ Citizenship

Cross-matching with state vital records databases holds significant promise as a method to document citizenship for U.S.-born individuals. The interim final regulations allow states to use a successful match with vital statistics data as documentation in lieu of an individual’s having to submit an original or certified birth certificate.[3] Most state’s have computerized vital records databases that contain records of certified births in the state, although it should be noted that state agencies cannot readily check automated records for state residents who were born in another state.

New analyses of Census data by the Center on Budget and Policy Priorities indicate that vital records matches can document the citizenship of the great majority of U.S.-born citizens. We examined data from the Census Bureau’s 2004 American Community Survey to determine the percentage of U.S.-born non-disabled individuals under age 65 with low incomes (below 150 percent of the poverty line) who live in the state where they were born.[4]

  • As Table 1 shows, 86 percent of low-income U.S.-born children and 67 percent of low-income U.S.-born adults live in the state where they were born and therefore could be identified through a match with state vital records.
  • These percentages vary widely across the states, but in every state, the great majority of native-born children and a substantial fraction of U.S.-born adults could be identified through a vital records match.

People for Whom State Vital Records Matches Could Work: Percentage of Low-income U.S.-Born Children and Adults Who Live n the State in Which They Were Born

State Children Under 18 Adults 18 to 64 State Children Under 18 Adults 18 to 64
US TOTAL 86% 67% Missouri 84% 62%
Alabama 89% 75% Montana 71% 54%
Alaska 77% 44% Nebraska 79% 66%
Arizona 78% 42% Nevada 59% 21%
Arkansas 76% 66% New Hampshire 65% 49%
California 95% 73% New Jersey 88% 71%
Colorado 78% 43% New Mexico 80% 62%
Connecticut 87% 57% New York 95% 82%
Delaware 77% 51% North Carolina 83% 68%
District of Columbia 90% 58% North Dakota 80% 67%
Florida 81% 44% Ohio 87% 76%
Georgia 85% 68% Oklahoma 75% 60%
Hawaii 86% 58% Oregon 77% 47%
Idaho 71% 41% Pennsylvania 89% 78%
Illinois 89% 76% Rhode Island 79% 59%
Indiana 88% 71% South Carolina 80% 66%
Iowa 74% 72% South Dakota 81% 59%
Kansas 76% 61% Tennessee 85% 63%
Kentucky 82% 73% Texas 92% 73%
Louisiana 93% 83% Utah 82% 55%
Maine 78% 63% Vermont 75% 46%
Maryland 84% 58% Virginia 75% 60%
Massachusetts 88% 71% Washington 80% 53%
Michigan 91% 79% West Virginia 78% 69%
Minnesota 74% 67% Wisconsin 78% 65%
Mississippi 87% 76% Wyoming 67% 42%
Notes: CBPP tabulations of the 2004 American Community Survey. Based on U.S.-born persons with incomes below 150 percent of the poverty line. Excludes those over the age of five who are disabled.

After accounting for the exemption from the new documentation requirement for Medicaid beneficiaries who also are enrolled in Medicare or SSI, we estimate that about 38 million U.S.-born individuals now enrolled in Medicaid will be subject to the new requirements when they seek to renew their coverage.[5] An additional 10 million American citizens are eligible apply for Medicaid each year; they, too, will be subject to the new requirements.[6] Administrative data indicate that more than half of the women enrolled because they have low incomes and are pregnant enter as new applicants during a year, as do one-quarter of children. As a result, the provisions that bar new applicants from receiving coverage until documentation is submitted could lead to delays in obtaining prenatal care for significant numbers of pregnant women and preventive and acute care for many children unless states institute systems to match records.

Cross-Matching With Other Databases Could Document Citizens’ Identities

Most Medicaid applicants and beneficiaries will be required to document their identity as well as their citizenship. Although a passport satisfies both requirements, a birth certificate counts only as evidence of citizenship, so individuals who present a birth certificate or have a vital records data match also must produce a separate document proving their identity.[7] This means that if a state opts to conduct a vital records data match but does not also conduct matches to obtain identity documents, individuals will still have to prove their identity by visiting the Medicaid office to present a driver’s license or other form of proof. That would compromise a state’s mail-in application and renewal procedures. Fortunately, the interim final regulations also allow states to document identity through cross-matches, using state or federal databases such as those used in the Food Stamp Program, the child support system, and motor vehicles departments.[8]

Using Food Stamp or TANF data may be particularly attractive, since Medicaid eligibility data systems often already are linked to eligibility data systems for state Food Stamp and TANF programs. Food Stamp regulations already require states to verify the identity of heads of household,[9] and states likely will already have gathered the identity documents of many low-income individuals in the process of documenting these individuals’ identity or other information (such as age or residency) for Food Stamps or TANF. The many Medicaid beneficiaries who are currently enrolled or previously have been enrolled in these programs could have their identities readily documented in this manner. In some cases, records from other programs such WIC or child support also are available and could be used in a similar fashion.

Some Medicaid eligibility agencies also may be able to obtain online access to data on driver’s licenses or state-issued identification cards from state motor vehicle departments. If so, this could provide proof of identity for a majority of adult Medicaid beneficiaries, who are likely to have either a driver’s license or state-issued identification card.

How Some States Are Implementing Data Matches to Document Citizenship and Identity

States that are conducting data matches or plan to do so are employing a variety of approaches. The approaches used to match data for beneficiaries seeking to retain their Medicaid coverage may differ in some respects from the methods used for new Medicaid applicants. State agencies that determine Medicaid eligibility can usually search their agency records to obtain identity documentation submitted for other benefit programs, but they must search records outside their own agency to document citizenship. Described below are the approaches being used by four states: Washington, Louisiana, Iowa, and Utah.[10]

Washington: focus on matching current beneficiaries. Washington State plans to conduct a large, batch-style vital records match of all Medicaid beneficiaries who are subject to the documentation requirement. In a batch-style match, data about Medicaid beneficiaries are matched against vital records data for thousands of records at a time rather than on a case-by-case basis.

In Washington State, the state Medicaid office will match hundreds of thousands of computerized Medicaid records against the vital records database. The state anticipates that the system will be very efficient, with a single programmer able to check about 35,000 records per day. Ten new staff will be on hand to search the vital records system further in cases where initial attempts at matching are unsuccessful. The state decided to conduct the matches centrally in order to avoid burdening eligibility workers with the job of doing the “detective work” necessary when a match is not readily available.

A positive match (i.e., proof that a Medicaid beneficiary was born in Washington State) will be recorded in the computerized Medicaid eligibility files, which are accessible by local eligibility offices. If a beneficiary’s citizenship cannot be ascertained through the vital records match, local eligibility staff will ask the beneficiary to submit paper documentation. In the event that beneficiaries are unable to submit such documentation, a centralized staff unit will attempt to obtain appropriate documentation for the person.

Washington also is developing approaches to conduct data matches for new applicants. State eligibility staff currently have the ability to do on-line “look-ups” of new applicants’ vital records, but this capability is somewhat more labor intensive as it requires eligibility staff to log in to a separate system. The state is trying to determine if it is possible to make this capability easier.

Louisiana: conducting “real-time” matches for applicants and beneficiaries. In Louisiana, local eligibility staff will conduct “real-time” online matches with vital records data as part of the normal process of determining Medicaid eligibility for new applicants and beneficiaries seeking to renew their coverage. “Real-time” matches allow eligibility workers to obtain the most up-to-date information available at any given moment.

Since Louisiana requires child applicants for Medicaid to provide proof of age, the state already has systems in place to permit a limited number of eligibility workers to look up vital records data on their computers and determine immediately whether there is a birth certificate in the state’s system. The match takes less than a minute. When a match is successful, the caseworker prints the screen and scans it into the electronic case record. To implement the new Medicaid documentation requirement, the state has expanded the authorization for access to the online vital records system to more than 1,000 Medicaid eligibility staff statewide.

One limitation of Louisiana’s system is that records for people born before 1984 are frequently unavailable since most older birth records have not been entered into the automated system.[11] Individuals for whom a match cannot be found will be asked to submit paper documentation. Louisiana also is checking available databases (including Food Stamps, TANF, Department of Labor, Social Security databases, like SDX or BENDEX, and WIC) to obtain proof of applicants’ and beneficiaries’ identities.

When proof of citizenship and identity has been obtained and scanned into the case record, the information is added to the Medicaid “person record,” along with a notation explaining how the information was verified. The state plans to keep the “person record” permanently so individuals who leave Medicaid and later return to the program do not have to re-supply citizenship and identity documentation.

Iowa: using a web-based request system. The Iowa Medicaid agency is very interested in conducting vital records data matching, but technical problems prevent the Medicaid staff from directly accessing the state’s vital records systems. To surmount this obstacle, the vital records and Medicaid agencies have developed a web-based system that allows Medicaid eligibility workers to request birth certificates on behalf of an applicant or current beneficiary.

The process works as follows. First, the eligibility worker enters the client’s Medicaid number into a web-based system designed for this purpose. The Medicaid eligibility system calls up information it has for that individual that could help in securing a vital records match, such as name, date of birth, and gender. (The eligibility worker can contact the client for additional information that could aid in finding a match — such as first, middle, and last name, county of birth, or mother’s maiden name — and then enter this information into the database.) The eligibility worker also indicates whether the request is for a new applicant or a current beneficiary. Because the CMS regulations require delaying coverage of new applicants until documentation is found, new applications receive priority.

Each night the system delivers a batch of requests to the vital records department, which looks up each request individually. When a match is found, vital records staff print a confirmation that the birth certificate is on file; this information is then provided to the Medicaid eligibility worker and becomes part of the person’s case record. If the vital records worker finds a possible (as opposed to a definite) match, he or she asks the Medicaid eligibility worker to get more information from the client so the match can be attempted again. As of this writing, Iowa was field testing this data-matching system; the time and effort involved in executing the matches are not yet known.

Utah: checking multiple databases using a single system. Established in 2004 to help reduce the state’s food stamp error rate, Utah’s “eFind” system has significantly improved the ease and efficiency with which caseworkers can search multiple databases to document individuals’ eligibility for a variety of public benefit programs. Previously, an eligibility worker had to log in to each separate database to obtain information. Now, some 18 sources of information, including databases and data warehouse files for Social Security, alien registration, motor vehicles, food stamps, TANF, and vital statistics, can be accessed through one system. The time needed to conduct a full search for information has been cut from 17 minutes to three.

Another advantage of eFind is that all caseworkers have access to it. Previously, only one or two caseworkers per eligibility team were authorized to use the vital statistics software. Now that birth records are contained in eFind, all caseworkers have access, and the vital records agency has been freed from the administrative tasks associated with authorizing individual caseworkers to use the system.

Utah is using eFind to implement the new Medicaid citizenship documentation requirement. After a caseworker indicates that a birth records search is desired for a given person, a check mark appears on the screen to indicate if a match exists. If no check mark appears, the caseworker can enter the birth records system to add information (such as mother’s maiden name) to attempt to secure a match. If a match is found, the caseworker prints out the screen and places it in the Medicaid case file.

Utah also has a separate records imaging system that scans and stores all documents submitted to substantiate a person’s eligibility for public benefits, including documents to prove income, immigration status, U.S. citizenship, and identity. The Departments of Workforce Services and of Health — the two agencies that determine Medicaid eligibility — are using this system to help Medicaid applicants and beneficiaries comply with the requirement to prove identity. For example, if a food stamp recipient applies for Medicaid, the caseworker searches the records imaging system to ascertain whether an image of the person’s driver’s license (or other identity document) is on file, in which case the individual does not need to produce that document a second time. The Department of Workforce Services is discussing including driver’s licenses in eFind so that system could be used to help document the identity of Medicaid applicants and beneficiaries, as well as their citizenship.

Addressing Technical and Logistical Challenges

States face a number of technical and logistical issues as they prepare to conduct data matches with vital records and other databases. Below are suggestions for dealing with some of the most common challenges.

  • Reduce administrative burden through interagency collaboration. The new Medicaid documentation requirement results in additional workload for both Medicaid and vital records agencies. This workload can be reduced through collaboration. While vital records offices benefit by collecting fees from individuals requesting birth certificates, few if any are equipped to handle the expected influx of requests by Medicaid applicants and beneficiaries. A strong case thus can be made that both the Medicaid agency and the vital records agency will benefit from the cross-matching of records.

    State Medicaid agencies can help pay the costs associated with cross-matching, either by paying a user fee for each vital record match or by covering the cost of the additional state staff involved in processing requests for documentation. These costs are eligible for reimbursement by the federal government at the Medicaid administrative matching rate of 50 percent.[12]

  • Engage technical experts within government agencies or from outside organizations. Some states may have difficulty identifying technical staff who have the knowledge and skills necessary to set up a system for effective data matching. Computer and programming staff are often in short supply. Sometimes, however, researchers from universities or health-related organizations have experience matching Medicaid and vital records data.

    For example, some researchers have developed this area of knowledge in carrying out research on birth outcomes as part of evaluations they are conducting of maternal and child health programs. Medicaid and other human service agencies also typically conduct other automated matching, such as matches with wage data from the Department of Labor and matches with immigrant data from the Department of Homeland Security’s Systematic Alien Verification of Entitlements (SAVE) database. These staff or private contractors may be able to help with matching.

  • Ensure that sufficient data are collected to execute efficient matches. In general, Medicaid applications solicit applicants’ names, dates of birth, and Social Security numbers (SSNs). All of these information elements can be used to match against vital records data.[13] Additional information is sometimes needed, however, to distinguish records that are nearly identical or when the usual basic information is not in the system. If, for example, a child has not yet been named when the birth certificate is recorded, additional information (such as the mother’s name and the child’s gender, birth date and birth place) can help complete the match. A number of states have revised or are in the process of revising their Medicaid applications or related forms to ensure that they collect critical information needed to conduct a successful match.

    Similarly, finding birth records for adult women can sometimes be a problem because women’s names often change due to marriage. Several states are beginning to collect mothers’ maiden names on Medicaid applications to improve matching. Washington State found it could sometimes infer a mother’s maiden name from her child’s birth certificate, and then conduct a search for the mother’s birth certificate using her maiden name.

  • Use multiple approaches to increase successful data matches. Using multiple approaches to matching can increase the number of cases matched. For example, a child health research project conducted by Connecticut Voices for Children needed to match birth records and Medicaid enrollment data. More than 9,000 records were matched using a mother’s SSN and the child’s date of birth; an additional 457 records were matched using the mother’s name and date of birth when the SSN match did not work. Because numbers may be transposed or names misspelled, multiple matching approaches can improve success rates when one field does not match.[14]
  • Find alternatives when data entry of vital records is delayed or incomplete. It may take a month or more for a newborn’s birth records to enter the vital records database. However, a hospital record on hospital letterhead dated close to the date of birth may also meet the documentation requirement.[15]

    A variety of technical problems may confound data matches. For example, a state may not have automated birth records for certain years or counties, and upgrading the systems or databases could require significant time and resources. CMS recently announced the availability of “Medicaid transformation” grants, which do not require state matching funds. States could consider applying for such grants to help improve their matching systems.[16]

  • Develop ways to turn near matches into successful matches. Techniques exist to improve matching rates by overcoming non-matches that occur due to data entry errors, such as transposed numbers or letters and slight variations in the spelling of names. The National Association for State Public Health Statistics and Information Systems (NASPHSIS) is exploring the use of techniques that identify close (but not exact) matches, which can then be reviewed more closely. For example, there is one matching technique, called Soundex, that can help match names despite slight variations or errors in the spelling of names (e.g., “Smith” vs. “Smyth.”).[17]
  • Develop ways to conduct data matching across states. The barriers created by the Medicaid documentation requirement could be alleviated further if state Medicaid agencies could conduct data matches with vital records agencies in other states. NASPHSIS is partnering with a broad array of state and federal agencies to develop a national system, called Electronic Verification of Vital Events, that can check birth certificates across the nation. The system is not expected to be ready until 2008, but NASPHSIS is investigating whether it can be used to share information across a limited number of states before then.[18] In addition, some states are considering vital records data-sharing reciprocity agreements with other states, particularly bordering states. This would significantly expand the potential scope of vital records matches.

Ways to Improve Matching to Document Citizenship

While the interim final regulations offer some important options to states, they do not give states options to develop or use other innovative approaches to data matches, such as the following:

  • More than 40 percent of all births in the United States are paid for by Medicaid.[19] Medicaid claims data (or claims from Medicaid managed care organizations) could be used to demonstrate when these babies have been born in U.S. hospitals and thus are native-born citizens. CMS should consider giving states the option to use electronic claims data to document citizenship for infants.
  • The Social Security Administration maintains the NUMIDENT data system, which includes data on the place of birth for virtually all people with Social Security numbers and citizenship data for those who applied for a Social Security number after 1972. The Social Security Administration has not made these data available to states and it is not clear whether CMS has pursued discussions with SSA to make the data available. These data could help states considerably because they indicate place of birth (and often citizenship) no matter where a person was born. CMS regulations could be modified to let states to use these data if they eventually become available.
  • Information from the Department of Homeland Security’s Systematic Alien Verification of Entitlements (SAVE) system can be used to document that a person is a naturalized citizen, although it has some limitations.[20] This approach can be particularly valuable for those who cannot find their original Certificates of Naturalization, the principal form of documentation permitted by CMS for naturalized citizens. Those who cannot find their original certificate will otherwise have to pay a $220 replacement fee, and obtaining the replacement can take up to a year. In comparison, a SAVE search typically costs less than $1 and can be completed in less than a week. Medicaid agencies generally already participate in SAVE, so this process is readily available to them. CMS should allow this information to be used in documenting that an applicant or beneficiary is a naturalized citizen. Under the interim regulations, data from the SAVE system is not considered satisfactory proof of citizenship.

Given how information-rich our nation is, it is plausible that other data bases or approaches exist that also could help to document citizenship. The interim final regulations do not accord flexibility that would permit states to use other effective or efficient strategies.


The new Medicaid documentation requirement creates difficult obstacles for U.S. citizens seeking to secure or retain Medicaid coverage, as well as significant administrative burdens for states. It could add to state administrative costs and cause substantial numbers of Medicaid-eligible U.S. citizens to have their coverage delayed or denied.

States can mitigate the harmful effects of the new requirement by adopting an option provided under the interim final regulations to document individuals’ citizenship and identity through data matches with existing state databases. Data matching has the potential to document citizenship status for most children on Medicaid and for a large fraction of adults. There also are a number of ways in which states can broaden the scope or improve the effectiveness of data matching and thereby reduce burdens and costs both on state administrators and on Medicaid applicants and beneficiaries.

End Notes

[1] Judith Solomon and Andy Schneider, “HHS Regulation Focuses New Medicaid Documentation Requirement on U.S. Citizen Families,” Center on Budget and Policy Priorities, July 13, 2006. Leighton Ku, “Revised Medicaid Documentation Requirement Jeopardizes Coverage For 1 to 2 Million Citizens,” Center on Budget and Policy Priorities, July 13, 2006.

[2] 71 Federal Register 39214 (July 12, 2006).

[3] 42 CFR 435.407(b)(1).

[4] We excluded elderly and disabled people since the Medicaid citizenship documentation rule effectively exempts Medicare and SSI beneficiaries. Like any survey data, these estimates are subject to sampling error. We thank Arloc Sherman of CBPP for tabulating these data.

[5] See Ku, July 13, 2006, for a discussion of the data and methodology. Some of the 38 million current beneficiaries will not seek renewal because they no longer have low incomes, gained private insurance, or have other reasons.

[6] This estimate is based on two analyses. CBPP analysis of 2003 Medical Expenditure Panel Survey data indicated that about 24 percent of those enrolled in Medicaid over the course of a year entered Medicaid after the year began; non-elderly adults and children are the most likely to enter. Unpublished analyses by Urban Institute of the 2003 Medicaid Statistical Information System data yielded similar results: 26 percent of children and adults on Medicaid were found to have entered the program after the first month of the year (personal communication with UI researchers, August 2006).

[7] The regulations give states the option of using matches with the State Data Exchange (SDX) — data from the Social Security Administration about SSI recipients — to document the citizenship and identity of SSI beneficiaries. This is most relevant in states that do not automatically provide Medicaid eligibility to those on SSI: Connecticut, Hawaii, Illinois, Indiana, Minnesota, Missouri, New Hampshire, North Dakota, Ohio, Oklahoma, and Virginia. States already routinely receive and use SDX data, so this should not be difficult for most states. Since the remaining states automatically grant Medicaid to those receiving SSI benefits, SSI recipients in those states are already exempt from the citizenship documentation requirements.

[8] 42 CFR 435.407(e)(10). The regulations make the Medicaid agency responsible for the accuracy of this information.

[9] 7 CFR 273.2 (f)(1)(vii).

[10] We spoke with Medicaid officials in those states in July and August 2006 and gratefully acknowledge their input.

[11] Plans to enter older data were delayed due to other priorities that have arisen since Hurricane Katrina. Fortunately, older records generally still exist because dedicated agency staff moved the records to secure locations before flooding occurred.

[12] State Medicaid Director Letter #06-012, June 9, 2006,, states on page 12 that federal Medicaid matching funds are available “for state expenditures to carry out the provisions of section 1903(x)” of the Social Security Act, where the new citizenship documentation requirement is set forth.

[13] A copy of the current U.S. standard certificate of live birth is at Actual forms used by states vary.

[14] Mary Alice Lee and Amanda Learned, “Linking Birth and Medicaid Data for MCH Policy and Program Development,” Connecticut Voices for Children, presentation at Association of Maternal and Child Programs conference, Feb. 2005.

[15] 42 CFR 435.407(c)(1).

[16] State Medicaid Director Letter #06-017, July 25, 2006, Enabling Medicaid and vital records data systems to interact would further the basic purpose of these grants, which is “to develop innovative methods to improve effectiveness and efficiency in providing medical assistance under Medicaid.”

[17] Soundex simplifies names by disregarding the vowels and comparing consonants. In addition, consonants that may sound alike (e.g., “c” and “k” or “s” and “z”) are grouped together. The National Archives and Records Administration has more information on the use of Soundex.

[18] Minnesota, Missouri, North Dakota, Iowa, and South Dakota are EVVE pilot states. Garland Land, the association’s director, can discuss possible projects with other states, using this platform ([email protected]). Land also notes that states could apply for Medicaid Transformation grants to upgrade their systems through the use of EVVE.

[19] Kaiser Family Foundation, “Births Financed by Medicaid as a Percent of Total Births, 2002.”

[20] Checking SAVE normally requires that the immigrant’s alien identification number or Certificate of Naturalization number be submitted to check for a match, but matches can be conducted even if these data are not available (although doing so is more difficult and less likely to be effective). The online system does not include information on naturalizations that occurred before 1975, although DHS has the ability to search older records when necessary.