The Social Security Administration’s Rejection of State Electronic Death Registration Reports

Audit Report

The Social Security Administration’s

Rejection of State Electronic Death

Registration Reports

A-08-18-50499 | September 2020

MEMORANDUM

Date:

September 21, 2020

Refer To:

To:

The Commissioner

From:

Inspector General

Subject:

The Social Security Administration’s Rejection of State Electronic Death Registration Reports

(A-08-18-50499)

The attached final report presents the results of the Office of Audit’s review. The objectives

were to assess the Social Security Administration’s rejection of State-submitted Electronic Death

Registration reports and determine whether these rejections resulted in the Agency

(1) improperly paying deceased beneficiaries and representative payees and (2) not posting death

information to the Numidents of deceased non-beneficiaries.

If you wish to discuss the final report, please call me or have your staff contact

Michelle L. Anderson, Assistant Inspector General for Audit, at 410-965-9700.

Gail S. Ennis

Attachment

The Social Security Administration’s Rejection of State

Electronic Death Registration Reports

A-08-18-50499

September 2020 Office of Audit Report Summary

Objectives

To assess the Social Security

Administration’s (SSA) rejection of

State-submitted Electronic Death

Registration (EDR) reports and

determine whether these rejections

resulted in the Agency (1) improperly

paying deceased beneficiaries and

representative payees and (2) not

posting death information to the

Numidents of deceased non-

beneficiaries.

Background

The EDR process begins when a

funeral director submits a decedent’s

personally identifiable information to

SSA’s Online Verification System

(OVS) through a State’s EDR

software. OVS determines whether the

information matches SSA’s Numident.

Once States certify the deaths, they

transmit the EDR reports to SSA.

SSA’s Death Information Processing

System (DIPS) performs a series of

verification checks on States’ EDR

reports. If the EDR report passes these

tests, SSA posts the death information

to its Numident and terminates

benefits. DIPS rejects EDR reports

that do not pass its verification tests.

From November 2017 through October

2018, SSA received 2,526,496 State-

submitted EDR reports. DIPS rejected

205,547 (8 percent). Of these, we

reviewed 13,989 (7 percent) that

contained personally identifiable

information that did match the

Numident but did not have a date of

death recorded on the Numident.

Findings

While DIPS rejects EDR reports that do not pass its verification

tests to prevent posting erroneous death information, SSA’s

rejection of 13,989 State-submitted EDR reports resulted in SSA

 improperly paying an estimated $36 million to 3,120 deceased

beneficiaries and 20 deceased representative payees and

 not posting dates of death to the Numidents of 10,849 deceased

non-beneficiaries.

We estimate identifying and correcting the missing dates of death

prevented, or will prevent, approximately $23 million in additional

improper payments to deceased beneficiaries and payees over the

next 12 months.

Of the 13,989 rejected EDR reports, DIPS rejected 13,829

(99 percent) because State records did not match exactly with the

most recent information in SSA’s OVS file. For example, DIPS

rejected one EDR report because it listed the decedent’s middle

name as “Lynne,” but the OVS file listed it as “Lynn.” We

matched the 13,829 rejected EDR reports with the Numident and

determined SSA would have posted correct death information had it

processed these EDR reports. Had States followed SSA

instructions to submit EDR reports that matched exactly with the

OVS file, DIPS would have posted the deaths. SSA could have

also prevented the improper payments had it designed DIPS to

perform a Numident match before it rejected EDR reports with

OVS discrepancies.

DIPS rejected the remaining 160 (1 percent) EDR reports because

they did not pass its other verification tests. For example, they

were missing death certificate numbers or had invalid dates. States

would need to correct and re-submit these EDR reports.

Recommendations

We made three recommendations for SSA to reduce improper

payments, correct its death records, and enhance the EDR process.

SSA agreed with our recommendations.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499)

TABLE OF CONTENTS

Objectives ........................................................................................................................................1

Background ......................................................................................................................................1

Results of Review ............................................................................................................................3

Impact of Rejected EDR Reports ...............................................................................................4

Causes of Rejected EDR Reports ..............................................................................................6

Conclusions ......................................................................................................................................7

Agency Actions Resulting from the Audit .......................................................................................8

Recommendations ............................................................................................................................9

Agency Comments ...........................................................................................................................9

– The Electronic Death Registration Process......................................................... A-1

– Electronic Death Registration Reports Transmitted from 2016 Through 2019 .. B-1

– Scope and Methodology ..................................................................................... C-1

– Sampling Methodology and Results ................................................................... D-1

– Agency Comments ............................................................................................... E-1

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499)

ABBREVIATIONS

C.F.R. Code of Federal Regulations

DIPS Death Information Processing System

EDR Electronic Death Registration

OIG Office of the Inspector General

OVS Online Verification System

PII Personally Identifiable Information

POMS Program Operations Manual System

SSA Social Security Administration

SSN Social Security Number

U.S.C. United States Code

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 1

OBJECTIVES

Our objectives were to assess the Social Security Administration’s (SSA) rejection of State-

submitted Electronic Death Registration (EDR) reports and determine whether these rejections

resulted in the Agency (1) improperly paying deceased beneficiaries

and representative payees

and (2) not posting death information to the Numidents

of deceased non-beneficiaries.

BACKGROUND

To identify and prevent payments after death, the Social Security Act requires that SSA establish

a program under which States can voluntarily provide SSA death data to match against its

Numident records.

SSA and the States developed the EDR program in 2002

to improve the

accuracy and timeliness of death information, reduce costs associated with sharing such

information, and reduce payments to deceased beneficiaries. SSA contracted with the National

Association for Public Health Statistics and Information Systems, a nonprofit organization that

represents States’ vital records and public health statistics offices, to develop nation-wide EDR

standards and guidelines. The Agency began processing States’

EDR reports in 2004. SSA

pays States to encourage them to submit EDR reports as soon as possible.

We use the term “beneficiary” throughout this report in reference to Old-Age, Survivors and Disability Insurance

beneficiaries and/or Supplemental Security Income recipients in current payment status.

Representative payees receive and manage the benefit payments of beneficiaries who cannot manage or direct the

management of their finances because of their age or mental and/or physical impairments. Social Security Act,

42 U.S.C. §§ 405(j) and 1383(a)(2) (govinfo.gov 2018).

The Numident is a database that stores personally identifiable information for all Social Security numberholders.

We use the term “non-beneficiary” throughout this report in reference to individuals who were not in current

payment status in December 2018.

Social Security Act, 42 U.S.C. § 405(r)(1) (govinfo.gov 2018).

Assessing Efforts to Eliminate Improper Payments: Hearing Before the Subcommittee on Federal Financial

Management, Government Information, Federal Services, and International Security, Senate Committee on

Homeland Security and Governmental Affairs, 112

Congress (2011) (statement of Kelly Croft, Deputy

Commissioner for Systems, Social Security).

As of the date of this review, 48 vital records jurisdictions, including New York City and the District of Columbia,

were participating in the EDR process. However, participating vital records jurisdictions do not necessarily report

all deaths to SSA through EDR. SSA, OIG, State Use of Electronic Death Registration Reporting, A-09-15-50023,

pp. C-1 and C-2 (July 2017). To simplify, we refer to vital records jurisdictions as States.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 2

The EDR process begins when a funeral director

submits a decedent’s personally identifiable

information (PII)—Social Security number (SSN), name, date of birth, and gender—to SSA’s

Online Verification System (OVS) through a State’s EDR software.

OVS determines whether

the decedent’s PII matches SSA’s Numident. OVS returns a “verified” or “unverified” response

to the funeral director, and SSA stores the submission and response in a data file. Funeral

directors provide States’ Bureaus of Vital Statistics with the (1) death information, (2) OVS

verification response (match or no match to SSA’s Numident), and (3) verification date for each

decedent. Once States’ Bureaus of Vital Statistics certify the death information from funeral

directors, the Bureaus transmit their EDR reports to SSA.

SSA’s Death Information Processing System (DIPS) conducts a series of verification checks on

States’ EDR reports to prevent posting erroneous death information to SSA records as well as

multiple payments to the States. For example, DIPS rejects EDR reports that are duplicates or

have missing, invalid, or improperly formatted data. DIPS also compares the EDR report it

receives against the most recent funeral director submission in the OVS file to ensure the two

records match exactly. If the name, date of birth, gender, OVS verification date, and OVS

response on the EDR report do not match exactly those in the most recent OVS submission,

DIPS rejects the EDR report.

Each year, SSA receives approximately 2.6 million State-submitted EDR reports. DIPS

processes approximately 92 percent of these EDR reports and rejects approximately 8 percent

that do not pass all its verification checks. For those DIPS processes, it automatically records the

death information on the Numident and terminates benefits, if applicable. If the decedent was

serving as a representative payee, SSA employees must replace the payee or send payments

directly to the beneficiary.

For those EDR reports DIPS rejects, SSA sends the States error

reports that contain codes that explain why DIPS rejected the reports. SSA encourages States to

correct and re-submit rejected EDR reports according to the instructions in its Death Process

Notice Guide.

See Appendix A for a flow chart of the EDR process. See Appendix B for EDR

Statistics for Calendar Years 2016 through 2019.

Hospital employees, coroners, medical examiners, and others with access to a State’s EDR software may also

verify the decedent’s information with SSA.

OVS provides real-time verification of individual SSNs.

20 C.F.R. §§ 404.2050(d), 416.650(d), 404.2055, and 416.655 (govinfo.gov 2019). SSA may suspend payment

under 20 C.F.R. §§ 404.2011(b) and 416.611(b) if it finds paying the beneficiary directly would cause substantial

harm, and the Agency cannot find a suitable alternative representative payee before the next payment is due.

SSA, Death Process Notice Guide, p.6 (November 2017).

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 3

SSA received 2,526,496 State-submitted EDR reports between November 2017 and

October 2018.

DIPS rejected 205,547 (8 percent) of these EDR reports. We used SSA’s

Enumeration Verification System

to match the PII in the rejected EDR reports against the

Numident and determined that 161,489 (79 percent) matched. Of the 161,489, we excluded

147,500 (91 percent) that had a date of death posted to the Numident by December 2018 or

contained duplicate SSNs to obtain our population of 13,989

unique SSNs for review.

these, we identified

 3,120 deceased beneficiaries who were in current pay,

 20 deceased payees who were receiving payments on 22 beneficiaries’ behalf, and

 10,849 deceased non-beneficiaries.

In July 2019, we reviewed a random sample of 100 of the 3,120 deceased beneficiaries. We also

reviewed the 20 deceased payees and 10,849 non-beneficiaries (see Appendix C for our audit

scope and methodology and Appendix D for our sampling methodology and results).

RESULTS OF REVIEW

While DIPS rejects EDR reports that do not pass its verification tests to prevent posting

erroneous death information, SSA’s rejection of 13,989 State-submitted EDR reports resulted in

SSA

 improperly paying an estimated $36 million to 3,120 deceased beneficiaries and 20 deceased

representative payees

and

 not posting dates of death to the Numidents of 10,849 deceased non-beneficiaries.

We estimate identifying and correcting the missing dates of death prevented, or will prevent,

approximately $23 million in additional improper payments to deceased beneficiaries and payees

over the next 12 months.

SSA retains detailed information on rejected EDR reports for 1 year. We selected November 2017 as the starting

period because that was the earliest month for which SSA had complete data when we initiated our review.

SSA’s Enumeration Verification System verifies batches of SSNs. Similar to OVS, it matches the SSN, name,

gender, and date of birth to SSA’s Numident records. It also checks for a death indicator on the Numident.

The 13,989 EDR reports DIPS rejected matched the Numident but did not have a date of death on the Numident.

We did not determine whether the date of death posted to the Numident matched the EDR report or SSA issued

improper payments to the individuals after it rejected the 147,500 EDR reports we excluded from our review.

We assumed SSA would have issued one improper payment after death regardless of whether it processed the

initial EDR report. Therefore, we started our improper payment computations 2 months after the date of death.

We based this estimate on the assumption that conditions will remain constant for the next 12 months.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 4

Of the 13,989 rejected EDR reports, DIPS rejected 13,829 (99 percent) because State records did

not match exactly with the most recent information in SSA’s OVS file. For example, DIPS

rejected one EDR report because it listed the decedent’s middle name as “Lynne,” but the OVS

file listed it as “Lynn.” We matched the 13,829 rejected EDR reports with the Numident and

determined SSA would have posted correct death information had it processed these EDR

reports.

Had States followed SSA instructions to submit EDR reports that matched exactly

with the OVS file,

DIPS would have posted the deaths. SSA could have also prevented the

improper payments had it designed DIPS to perform a Numident match before it rejected EDR

reports with OVS discrepancies.

DIPS rejected the remaining 160 (1 percent) EDR reports because they did not pass its other

verification tests. For example, they were missing death certificate numbers or had invalid dates.

States would need to correct and re-submit these EDR reports.

Impact of Rejected EDR Reports

SSA improperly paid approximately $1.2 million to the 100 beneficiaries in our sample but had

recovered

approximately $200,000 (see Table 1). SSA subsequently recorded the deaths of the

33 beneficiaries with rejected EDR reports

based on death reports from non-EDR sources. It

took SSA an average of 229 days

to record the deaths and terminate payments after it rejected

the EDR reports. As of July 2019, SSA had suspended 11 beneficiaries’ payments, and the

remaining 56 were still in current pay status.

We used SSA’s Enumeration Verification System to match the PII on the EDR report to the Numident. We

obtained death certificates for some sampled beneficiaries and representative payees and determined the PII matched

the EDR report. However, we did not obtain death certificates for all the individuals in our population.

SSA, Death Process Notice Guide, p.5 (November 2017).

States would have been unable to successfully re-submit 5 (3 percent) of the 160 EDR reports because SSA

previously paid the States for their death reports. However, these decedents’ Numidents did not contain a date of

death.

SSA recovers improper payments after death through the Department of the Treasury reclamations, refunded

benefits, and/or Medicare premiums. SSA, POMS, GN 02408.610, A.3 (September 3, 2020), GN 02210.150, A

(August 16, 2006), and GN 02330.100 (December 22, 1989).

SSA recorded incorrect dates of death for 4 of the 33 beneficiaries. The incorrect dates of death resulted in

improper payment computations and recovery actions for two of the four beneficiaries.

The mean was 229 days, and the median was 198 days.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 5

Table 1: Results of 100 Sampled Beneficiaries

Status

Number of

Beneficiaries

Improper

Payments

Improper

Payments

Recovered

Payments Terminated and Death Recorded

on the Numident

33 $274,998 $176,163

Payments Suspended but Death Not

Recorded on the Numident

11 100,448 30,014

In Current Pay and Death Not Recorded on

the Numident

56 798,995 0

Total

100

$1,174,441

$206,177

Of the 20 representative payees, SSA subsequently removed 11 who were serving as payees for

13 beneficiaries. However, SSA issued $84,615 to the nine payees who remained on its records.

If a representative payee dies and is not replaced, SSA cannot be sure the funds are being used to

provide the beneficiary food, clothing, shelter, and medical care.

If another individual acts as a

representative payee without SSA’s knowledge, that person is not subject to SSA’s financial

oversight and reporting responsibilities required by a representative payee—such as reporting

events to SSA that may affect the individual’s entitlement or benefit payment amount.

Of the 10,849 non-beneficiaries, SSA had recorded the death of 1,032 (10 percent) on the

Numident by July 2019. SSA would have recorded these deaths, on average, 233 days sooner

had it processed the rejected EDR reports. While SSA is not issuing improper payments to the

non-beneficiaries, it uses death information on the Numident to create a file of deaths it shares

with other Federal benefit-paying agencies.

Therefore, resolving discrepancies can improve the

accuracy and completeness of death information it shares with these Federal benefit-paying

agencies.

By January 2020, SSA had posted the death for an additional 230 non-beneficiaries,

leaving 9,587 non-beneficiaries

without a date of death on their Numidents.

20 C.F.R. §§ 404.2040(a) and 416.640(a) (govinfo.gov 2019).

20 C.F.R. §§ 404.2035 and 416.635 (govinfo.gov 2019).

The mean was 233 days, and the median was 209 days.

Other Federal benefit-paying agencies include the Railroad Retirement Board, Centers for Medicare and Medicaid

Services, Internal Revenue Service, Department of Veterans Affairs, and Office of Personnel Management.

Although SSA shares its death information with other Federal benefit-paying agencies, those agencies should

independently verify the individual’s death before they take adverse action.

Of the 10,849 non-beneficiaries, SSA posted the death for 1,032 by July 2019 and an additional 230 by

January 2020. Therefore, 9,587 non-beneficiaries did not have a date of death on their Numidents.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 6

Causes of Rejected EDR Reports

Of the 13,989 rejected EDR reports, DIPS rejected 13,829 (99 percent) because they did not

match exactly the most recent funeral director submission on SSA’s OVS file. We manually

reviewed the OVS discrepancies in our sample of beneficiaries in current pay

and determined

SSA rejected 25 percent of the EDR reports because the OVS verification date on the EDR

report did not match the OVS file. SSA rejected other sampled EDR reports because the date of

birth, OVS response, or name (first, middle, or last) was not an exact match to the OVS file.

For example, SSA rejected one EDR report because it listed the decedent’s last name as

“O’Connell,” but OVS listed it as “Oconnell.”

SSA designed the OVS verification check in DIPS to prevent erroneous death postings.

However, SSA did not include tolerances for minor discrepancies when it designed this check

thereby preventing some EDR reports with correct death information from being posted. We

found SSA had not studied the effectiveness of EDR rejections or changed the EDR rejection

process before our review.

SSA rejected 123,632 (60 percent) of the 205,547 EDR reports in

our population because of the OVS verification check even though the PII on the EDR report

matched the Numident. Of the 123,632 EDR reports, 13,829 deceased individuals did not have

death information posted to their Numidents as of December 2018.

Had SSA processed these

EDR reports, it would have posted correct death information.

Had States followed SSA

instructions to submit EDR reports that matched exactly with the OVS file,

DIPS would have

posted the deaths. SSA could have also prevented the improper payments had it designed DIPS

to perform a Numident match before rejecting EDR reports with OVS discrepancies.

Of our 100 sample, SSA rejected 99 because of OVS discrepancies and 1 because it did not contain a valid date of

death.

Some EDR reports contained discrepant information in multiple fields of the OVS file. For example, SSA

rejected one record because both the OVS verification date and the middle name did not match.

In November 2019, SSA implemented changes to DIPS that reduced EDR rejections based on non-alphanumeric

discrepancies between the EDR report and OVS. SSA implemented this change as a result of our review and it

partially corrects our findings for deaths reported after November 2019.

We excluded 109,803 rejected EDR reports from our review because they contained duplicate SSNs or had death

information posted to their Numidents as of December 2018. We did not determine whether SSA’s rejection of

these EDR reports resulted in delayed death postings or improper payments.

We used SSA’s Enumeration Verification System to match the PII on the EDR report to the Numident. We

obtained death certificates for some sampled beneficiaries and representative payees and determined the PII matched

the EDR report. However, we did not obtain death certificates for all the individuals in our population.

SSA, Death Process Notice Guide, p.5 (November 2017).

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 7

DIPS rejected the remaining 160 (1 percent) EDR reports because they did not pass its other

verification checks. For example, they were missing death certificate numbers or had invalid

dates. SSA rejected one EDR report because it showed May “00” as the date of death. While

these were State-certified deaths, States would have to correct and re-submit these EDR reports

before DIPS could process them.

SSA provides States instructions for correcting and re-submitting rejected EDR reports along

with an email contact to request further information. SSA does not verify that States

successfully re-submit rejected EDR reports. The seven State officials with whom we spoke

stated they did not review the EDR rejection reports or understand why SSA rejected the EDR

reports. Several State officials told us they did not understand how their software could submit

EDR reports that contained discrepant information from that on the OVS file.

In fact, of the

13,989 rejected EDR reports, only 1 State successfully re-submitted 29 of the previously rejected

reports. Without State correction and re-submission, when DIPS rejects an EDR report, SSA

must rely on information it receives from non-EDR sources, such as relatives, to post the death.

CONCLUSIONS

Of the approximately 2.6 million EDR reports States submitted in 2019, DIPS successfully

processed 2.4 million (92 percent). While DIPS is working as designed, SSA could have

avoided paying approximately $36 million to the 3,120 deceased beneficiaries and

20 representative payees in our review had DIPS processed, instead of rejected, the State-

submitted EDR reports. Several of our prior reviews,

covering 1968 through 2018, identified

State death information that did not appear on SSA’s Numident and/or payment records. This

discrepant death information resulted in SSA improperly paying $113 million to deceased

beneficiaries across 13 States.

We determined SSA could improve the EDR process and reduce

discrepancies between SSA and State death records by including a Numident match in DIPS to

assess OVS discrepancies before it rejects EDR reports. Enhancing DIPS would also improve

the accuracy and timeliness of SSA’s death records and reduce improper payments after death.

States would have been unable to successfully re-submit 5 (3 percent) of the 160 EDR reports because SSA

previously paid the States for their death reports. However, these decedents’ Numidents did not contain a date of

death.

Although some of the 48 States participating in the EDR process developed their own death registration systems,

most States acquired EDR software from vendors.

SSA, OIG, Match of Maine and Rhode Island Death Data Against SSA Records, A-01-18-50314, p. 2 and

Appendix A (September 2019).

We included Puerto Rico in the list of States we reviewed even though Puerto Rico is a U.S. Territory. We did

not include Department of the Treasury, Department of Veterans Affairs, or Office of Personnel Management in the

list of States or the $113 million. Not all the unrecorded deaths identified in our prior reviews occurred in

participating EDR States or during periods that EDR was available.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 8

AGENCY ACTIONS RESULTING FROM THE AUDIT

In September 2019, we provided SSA the death certificates for the sampled beneficiaries with

missing or incorrect death information as well as the EDR information for the current pay cases.

We requested that SSA take appropriate action to terminate benefits and record the dates of death

on the Numidents of the remaining 3,091 beneficiaries in our population.

As of June 2020,

SSA had terminated benefits and recorded correct death information for 1,144 beneficiaries.

SSA continued issuing monthly payments to 26 of the deceased beneficiaries in our sample, and

1,947 cases from our population still require SSA action.

We provided SSA the death certificates for nine deceased representative payees. As of

January 2020, SSA had replaced seven of the deceased representative payees and determined

one beneficiary was capable of managing his/her own benefits. SSA stopped payments to the

remaining payee but has not selected a new payee or issued payments directly to the beneficiary

because SSA employees have not located the beneficiary. Since SSA took appropriate action,

we did not make a recommendation regarding the deceased representative payees.

For the 9,587 deceased non-beneficiaries, we requested SSA take necessary action to record the

deaths on the Numident. SSA advised us that, as the non-beneficiary cases did not involve

improper payments, any future cleanup of these cases is resource-dependent. SSA plans to post

death information for the non-beneficiary cases identified in this audit that pass its screening

process via its information technology modernization efforts in Fiscal Year 2021.

In November 2019, SSA implemented changes to DIPS that reduced EDR rejections based on

non-alphanumeric

discrepancies between the EDR report and OVS. For example, after this

change, DIPS would not reject an EDR report if it lists the decedent’s last name as “O’Connell,”

but OVS lists it as “Oconnell.” This change partially corrects our findings for deaths reported

after November 2019. Had this change been in place at the time DIPS processed our

100 sampled beneficiaries, SSA could have avoided improper payments to 23 deceased

beneficiaries. SSA also advised us that it plans to begin analyzing improvements to the EDR

system through its information technology modernization initiative, depending on resources, in

Fiscal Year 2022.

SSA correctly posted the deaths of 29 sampled beneficiaries. Therefore, of our population of 3,120 beneficiaries,

only 3,091 require that SSA record their death on the Numident. The EDR reports for four beneficiaries did not

contain valid dates of death and will require that SSA employees develop the correct date of death.

Of the 1,144, SSA advised us it corrected the date of death for 4 beneficiaries based on our audit. We do not

know how many of the remaining 1,140 death postings resulted from our audit and how many resulted from other

death reports.

Of the 1,947 cases that require SSA action, 46 contain discrepant death information on the EDR report and

Numident.

Examples of non-alphanumeric discrepancies include spaces, hyphens, and apostrophes that appear in the EDR

report but not in SSA’s OVS.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) 9

RECOMMENDATIONS

We recommend that SSA:

1. Record deaths on the Numident, terminate payments, and collect the overpayments for the

1,947 remaining deceased beneficiaries.

2. Add death information to the Numident, as appropriate, for the 9,587 remaining deceased

non-beneficiaries.

3. Include a Numident match in DIPS, or similar process, to assess OVS discrepancies before

rejecting an EDR report.

AGENCY COMMENTS

SSA agreed with our recommendations. The full text of SSA’s comments is included in

Appendix E.

Michelle L. Anderson

Assistant Inspector General for Audit

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499)

APPENDICES

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) A-1

– THE ELECTRONIC DEATH REGISTRATION

PROCESS

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) B-1

– ELECTRONIC DEATH REGISTRATION REPORTS

TRANSMITTED FROM 2016 THROUGH 2019

States’ Bureau of Vital Statistics

use the Electronic Death Registration (EDR) process to

transmit death information to the Social Security Administration (SSA). SSA began processing

States’ EDR reports in 2004. See Table B–1 for EDR statistics in Calendar Years 2016 through

2019.

Table B–1: Electronic Death Registration Statistics for 2016 Through 2019

Calendar

Year

State EDR Reports

Transmitted to SSA

EDR Reports

Rejected by SSA

Rejection

Percentage

Acceptance

Percentage

2016

2,179,343

204,672

9.4

90.6

2017

2,477,293

206,389

8.3

91.7

2018

2,609,935

208,090

8.0

92.0

2019

2,662,265

222,485

8.4

91.6

TOTAL

9,928,836

841,636

8.5

91.5

Source: SSA’s Death Management Information System

As of the date of this review, 48 vital records jurisdictions, including New York City and the District of Columbia,

were participating in the EDR process. However, participating vital records jurisdictions do not report all deaths to

SSA through EDR. SSA, OIG, State Use of Electronic Death Registration Reporting, A-09-15-50023, pp. C-1 and

C-2 (July 2017). For simplification purposes, we refer to vital records jurisdictions as States.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) C-1

– SCOPE AND METHODOLOGY

To accomplish our objectives, we:

 Reviewed pertinent sections of the Social Security Administration’s (SSA) policies and

procedures, applicable laws, and regulations.

 Reviewed prior Office of the Inspector General and Government Accountability Office

reports.

 Interviewed officials from the National Association for Public Health Statistics and

Information Systems as well as representatives from vital records offices in seven States.

 Reviewed information from SSA’s Death Management Information System, Representative

Payee System, Master Beneficiary Record, Payment History, Supplemental Security Income

Record, and Numident.

 Obtained a file of 205,547 Electronic Death Registration (EDR) reports that SSA’s Death

Information Processing System rejected between November 1, 2017 and October 22, 2018.

We submitted the information from the rejected EDR reports through SSA’s Enumeration

Verification System

to determine whether (1) the personally identifiable information on the

EDR report matched SSA’s Numident and (2) SSA had recorded death information on the

Numident. We identified 13,989 deceased individuals with rejected EDR reports whose

personally identifiable information matched SSA records but did not have a date of death on

SSA’s Numident in December 2018.

Of the 13,989, we identified

 3,120 deceased individuals in current payment status. We selected a random sample of

100 Social Security numbers and determined the following as of July 2019.

 Whether SSA posted the correct date of death to the Numident.

 Whether SSA posted the death information based on an EDR or non-EDR source.

 The amount of improper payments SSA issued after the beneficiary’s death.

The Numident is a database that stores personally identifiable information for all Social Security numberholders.

SSA only retains detailed information on rejected EDR reports for 1 year. We selected November 2017 as the

starting period because that was the earliest month SSA had complete data when we initiated our review.

SSA’s Enumeration Verification System is an internal application that verifies batches of Social Security numbers.

Similar to SSA’s Online Verification System, it matches the Social Security number, name, gender, and date of birth

to SSA’s Numident records. It also checks for a death indicator on the Numident.

SSA receives death reports from non-EDR sources, such as relatives, representative payees, and signed statements

from funeral directors.

To be conservative, we assumed SSA would have issued one improper payment after death regardless of whether

the Agency processed the initial EDR report. Therefore, we started our improper payment computations 2 months

after the date of death.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) C-2

 Whether SSA had terminated or suspended benefits.

 The amount of improper payments SSA recovered.

 The number of days it took SSA to post death information to the Numident after it

rejected the EDR report (if applicable).

 Discrepant information between the EDR report and SSA’s Online Verification

System file.

 20 deceased representative payees. We determined the following as of July 2019.

 The number of beneficiaries represented by the deceased payee.

 Whether SSA posted the death information based on an EDR or non-EDR source.

 Whether SSA had removed the deceased payees from the beneficiaries’ records.

 The amount of improper payments SSA issued to deceased representative payees.

 10,849 deceased non-beneficiaries. We determined the number of individuals who had a

date of death posted to their Numidents as of July 2019 and January 2020. In addition,

we determined whether SSA posted the death based on an EDR or non-EDR source.

 We obtained death certificates for the deceased beneficiaries in our sample who had missing

or discrepant dates of death on SSA’s Numident. We also obtained death certificates for the

nine deceased representative payees who remained on SSA records as of July 2019.

 We provided SSA the death certificates we obtained. We also provided SSA the EDR

information for the remaining current pay and non-beneficiary cases. While we did not

identify specific instances of fraud, we provided the death information for the deceased

beneficiaries to our Office of Investigations.

See Footnote 5.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) C-3

We conducted our audit in Birmingham, Alabama, from August 2018 through February 2020.

We determined the computer-processed data were sufficiently reliable for our intended use. We

conducted tests to determine the completeness and accuracy of the data. These tests allowed us

to assess the reliability of the data and achieve our audit objectives.

We assessed the significance of internal controls necessary to satisfy the audit objective. This

included an assessment of the five internal control components, including control environment,

risk assessment, control activities, information and communication, and monitoring. In addition,

we reviewed the principles of internal controls as associated with the audit objective. We

identified the following three components and five principles as significant to the audit objective.

 Component 2: Risk Assessment

 Principle 6: Define objectives and risk tolerances

 Principle 7: Identify, analyze, and respond to risks

 Component 3: Control Activities

 Principle 11: Design activities for the information system

 Component 5: Monitoring

 Principle 16: Perform monitoring activities

 Principle 17: Evaluate issues and remediate deficiencies

The SSA entity audited was the Office of the Deputy Commissioner for Systems. We conducted

this performance audit in accordance with generally accepted government auditing standards.

Those standards require that we plan and perform the audit to obtain sufficient, appropriate

evidence to provide a reasonable basis for our findings and conclusions based on our audit

objectives. We believe the evidence obtained provides a reasonable basis for our findings and

conclusions based on our audit objectives.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) D-1

– SAMPLING METHODOLOGY AND RESULTS

In December 2018, we identified 3,120 beneficiaries

in current pay status whose personally

identifiable information from States’ Electronic Death Registration reports matched the

Social Security Administration’s (SSA) Numident,

but the Numident did not contain a date of

death. In July 2019, we reviewed a random sample of 100 beneficiaries (see Table D–1).

Table D–1: Population and Sample Size

Description

Deceased Beneficiaries in

Current Payment Status

Population

3,120

Sample Size

100

Based on our review, SSA issued $1,174,441 in improper payments after death to the

100 sampled beneficiaries. To estimate the improper payments for the 3,120 population, we

used a straight-line approach. We estimated SSA had issued approximately $36.6 million in

improper payments to the 3,120 deceased beneficiaries in our population.

Between the date SSA rejected the Electronic Death Registration reports and our review, SSA

had recovered approximately $6.4 million of these improper payments (see Table D–2).

Accordingly, we estimate SSA has yet to recover approximately $30.2 million in improper

payments issued to deceased beneficiaries in our population (see Table D–3).

We use the term “beneficiary” in reference to Old-Age, Survivors and Disability Insurance beneficiaries and/or

Supplemental Security Income recipients in current payment status.

The Numident is a database that stores personally identifiable information for all Social Security numberholders.

Straight-line estimation computations: $1,174,441.20 improper payments in sample ÷ 100 beneficiaries in sample

= $11,744.41; and then $11,744.41 × 3,120 beneficiaries in population = $36,642,565 (rounded) estimated improper

payments to beneficiaries in the population.

SSA recovers improper payments after death through the Department of the Treasury reclamations, refunded

benefits, and/or Medicare premiums. SSA, POMS, GN 02408.610, A.3 (September 3, 2020), GN 02210.150, A

(August 16, 2006), and GN 02330.100 (December 22, 1989).

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) D-2

Table D–2: Recovery of Improper Payments to Deceased Beneficiaries

Description

Number of

Beneficiaries with Full

or Partial Recovery

Improper Payments

Recovered

Sample Results

$206,177

Point Estimate

1,217

$6,432,731

Projection - Lower Limit

966

$4,443,241

Projection - Upper Limit

1,483

$8,422,222

Note: All statistical projections are at the 90-percent confidence level.

Note: For the 100 sampled deceased beneficiaries, SSA recovered all improper payments issued to 16.

SSA also recovered some, but not all, of the improper payments issued to 23.

Table D–3: Improper Payments to Deceased Beneficiaries Not Recovered

Description

Number of

Beneficiaries Without

Full Recovery

Improper Payments Not

Recovered

Sample Results

$968,264

Point Estimate

2,621

$30,209,834

Projection - Lower Limit

2,398

$26,107,069

Projection - Upper Limit

2,796

$34,312,600

Note: All statistical projections are at the 90-percent confidence level.

For the deceased beneficiaries, we further estimate identification and correction of the missing

death information prevented, or will prevent, approximately $23 million in additional improper

payments after death over the next 12 months (see Table D–4).

Table D–4: Improper Payments to Deceased Beneficiaries over the Next 12 Months

Description

Number of Deceased

Beneficiaries

Improper Payments

Avoided

Sampled Beneficiaries in Current

Payment Status as of July 2019

56 $738,013

Point Estimate

1,747

$23,026,012

Projection – Lower Limit

1,480

$19,121,221

Projection – Upper Limit

2,006

$26,930,802

Note: All statistical projections are at the 90-percent confidence level.

We based this estimate on the assumption that conditions will remain constant for the next 12 months.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) D-3

We also identified 20 deceased representative payees

who were receiving payments for

22 beneficiaries in December 2018. As of July 2019, SSA had issued the representative payees

$159,311 in improper payments after their deaths. SSA had removed 11 payees from beneficiary

records.

SSA also had paid $84,615 to the 9 payees who remained active on SSA records. SSA

had suspended payments to one of the beneficiaries with a deceased payee. Therefore, SSA was

only issuing ongoing improper payments to eight of the nine payees. Based on our review, SSA

took action to stop payments to the eight deceased payees. By taking action on these cases, we

estimate SSA will prevent $62,943 in additional payments after death over the next 12 months

(see Table D–5).

Table D–5: Improper Payments to Deceased Representative Payees

Description Amounts

Number of Deceased Representative Payees Who Received

Improper Payments

Total Payments Issued After Representative Payees’ Deaths

$159,311

Number of Deceased Representative Payees Who Remained

on Beneficiary Records in July 2019

Uncorrected Improper Payments to Deceased Representative

Payees Who Remained on SSA Records

$84,615

Number of Deceased Representative Payees With Ongoing

Improper Payments in July 2019

Estimated Payment After Death over the Next 12 Months

$62,943

Representative payees receive and manage the benefit payments of beneficiaries who cannot manage or direct the

management of their finances because of their age or mental and/or physical impairments. Social Security Act,

42 U.S.C. §§ 405(j) and 1383(a)(2) (govinfo.gov 2018).

We could not determine the amount of improper payments SSA recovered from these deceased payees.

We based this estimate on the assumption that conditions will remain constant for the next 12 months.

SSA’s Rejection of State Electronic Death Registration Reports (A-08-18-50499) E-1

– AGENCY COMMENTS

SOCIAL SECURITY

Office of the Commissioner

MEMORANDUM

Date:

September 10, 2020 Refer To: S1J-3

To:

Gail S. Ennis

Inspector General

From:

Stephanie Hall

Chief of Staff

Subject:

Office of the Inspector General Draft Report “The Social Security Administration’s Rejection of

State Electronic Death Registration Reports” (A-08-18-50499) -- INFORMATION

Thank you for the opportunity to review the draft report. We agree with the recommendations.

We are pleased the report acknowledges our efforts to maintain the accuracy of our records and

the effectiveness of the Electronic Death Registration (EDR) process. We remain vigilant when

posting death information to our records because inaccurate information can have adverse

consequences for affected individuals. Although the vast majority of EDR transactions are

accurate and timely, States report some deaths that do not pass our data reliability tests. To

ensure we do not post a death that did not occur, we return those EDR reports to the States for

correction.

We will continue to collaborate with States to expand EDR and provide training on accurately

submitting or resubmitting rejected information in EDR. Our ongoing efforts continue to

improve the accuracy and efficiency of our death reporting processes, promote program integrity,

and prevent improper payments.

Please let me know if we can be of further assistance. You may direct staff inquiries to

Trae Sommer at (410) 965-9102.

MISSION

By conducting independent and objective audits, evaluations, and investigations, the Office of

the Inspector General (OIG) inspires public confidence in the integrity and security of the Social

Security Administration’s (SSA) programs and operations and protects them against fraud,

waste, and abuse. We provide timely, useful, and reliable information and advice to

Administration officials, Congress, and the public.

CONNECT WITH US

The OIG Website (https://oig.ssa.gov/) gives you access to a wealth of information

approximately OIG. On our Website, you can report fraud as well as find the following.

• OIG news

• audit reports

• investigative summaries

• Semiannual Reports to Congress

• fraud advisories

• press releases

• congressional testimony

• an interactive blog, “

Beyond The

Numbers” where we welcome your

comments

In addition, we provide these avenues of

communication through our social media

channels.

Watch us on YouTube

Like us on Facebook

Subscribe to our RSS feeds or email updates

OBTAIN COPIES OF AUDIT REPORTS

To obtain copies of our reports, visit our Website at https://oig.ssa.gov/audits-and-

investigations/audit-reports/all. For notification of newly released reports, sign up for e-updates

at https://oig.ssa.gov/e-updates.

REPORT FRAUD, WASTE, AND ABUSE

To report fraud, waste, and abuse, contact the Office of the Inspector General via

Website: https://oig.ssa.gov/report-fraud-waste-or-abuse

Mail: Social Security Fraud Hotline

P.O. Box 17785

Baltimore, Maryland 21235

FAX: 410-597-0118

Telephone: 1-800-269-0271 from 10:00 a.m. to 4:00 p.m. Eastern Standard Time

TTY: 1-866-501-2101 for the deaf or hard of hearing