This is the accessible text file for GAO report number GAO-03-1113R entitled 'Social Security Administration: Subcommittee Questions Concerning Efforts to Automate the Disability Claims Process' which was released on September 05, 2003. This text file was formatted by the U.S. General Accounting Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. September 5, 2003: The Honorable E. Clay Shaw, Jr. Chairman, Subcommittee on Social Security Committee on Ways and Means House of Representatives: Subject: Social Security Administration: Subcommittee Questions Concerning Efforts to Automate the Disability Claims Process: Dear Mr. Chairman: This letter responds to your August 12, 2003, request that we provide answers to questions relating to our July 24, 2003, testimony.[Footnote 1] In that testimony, we discussed the risks that the Social Security Administration (SSA) faces in its efforts to automate its disability claims process. Your questions, along with our responses, follow. 1. The Social Security Administration (SSA) has indicated that the agency could potentially save $1 billion, at an estimated cost of approximately $900 million, by implementing an electronic disability folder. In your opinion, does the $900 million cost of this project appear to be a reasonable estimate? Is the $1 billion in savings a reasonable estimate? Is there a possibility that the cost of the project could balloon? Our work to date on SSA's February 2003 cost-benefit analysis raises concerns that SSA may have underestimated its accelerated electronic disability (AeDib) system costs. For example, the cost-benefit analysis did not fully consider the costs associated with certain critical information technology infrastructure elements supporting the nationwide rollout, such as scanning and imaging by the outsourced vendor, telecommunications, disaster recovery, and on-site retention and destruction of source documents, such as medical records. Because SSA has not yet fully estimated these costs, we are unclear about their magnitude. We are also concerned that the corresponding benefits cited in SSA's cost-benefit analysis may be overstated. Specifically, our review found that certain assumptions used in the analysis could be too optimistic. For example, SSA estimated benefits based on an assumption that state Disability Determination Services (DDS) would receive 30 percent of all medical evidence in electronic form by 2004. However, state DDS officials with whom we spoke contend that their offices currently receive about 11 percent or less of medical evidence electronically, and disagree with the 30-percent assumption. As to whether the costs of AeDib could balloon, the possibility of cost increases exists for any project the size and magnitude of AeDib. As mentioned previously, the existing estimates do not include certain costs, such as some costs for outsourced scanning and imaging and the cost of disaster recovery. Including these could add to SSA's overall cost estimate. Our work analyzing the costs and benefits of AeDib is ongoing; the final results will be included in our report to be issued to you later this year. 2. According to your testimony, the SSA has developed a risk management plan, but is still without a comprehensive assessment of risk that could affect the electronic disability folder. Can you explain the difference between the risk management plan and a comprehensive assessment of risks? How does their risk management plan fall short of what GAO recommends? Why is this plan so important? A risk management plan provides guidance to project management teams and requires them to proactively identify facts and circumstances that could increase the probability of failing to meet project commitments, and take steps to prevent this from occurring.[Footnote 2] A comprehensive assessment of risks, which is completed according to the risk management plan, is the process of identifying risks with a high probability and cost of failure, and developing strategies for mitigating those risks.[Footnote 3] Based on our work to date, we do not have concerns about SSA's risk management plan, which was developed in accordance with our own and Office of Management and Budget (OMB) guidance. However, as mentioned in our testimony, we were concerned that SSA had not completed a subsequent comprehensive risk assessment. SSA officials agreed with the need for a comprehensive risk assessment, and since our testimony, the agency has provided us with detailed risk assessments for four of the five AeDib projects--the Electronic Disability Collect System, the Document Management Architecture, a DDS systems migration and electronic folder interface, and Internet disability applications. SSA has not yet provided us with a time frame for completing the remaining risk assessment for its Hearings and Appeals case processing management system. A risk management plan is an essential tool used to guide the development of a comprehensive assessment of risks and mitigation strategies. A comprehensive risk assessment is equally critical. Both tools can help SSA avoid potential problems before they manifest themselves in cost, schedule, and performance shortfalls. 3. The GAO has identified a number of areas where the SSA could improve. For instance, the agency needs to develop risk assessment tools and could take additional measures to involve key stakeholders in the systems development process. Is there any indication that the SSA is acting on your recommendations? SSA has acknowledged our concerns relative to the areas of improvement we identified in our testimony, and has taken some action to address them. However, more work remains to fully address these issues. Specifically, we expressed the need for SSA to perform a comprehensive risk assessment to identify project risks and establish mitigation strategies for them. As noted in our response to question 2, SSA recently provided us with detailed risk assessments for four of the five AeDib projects, but has not yet provided a detailed assessment of risks for its remaining project. We also commented on the need for SSA to perform end-to-end testing prior to implementation, to ensure that the system it is developing will perform as intended. To date, SSA has not yet finalized its test and evaluation strategy. Therefore, it is unclear whether SSA intends to include end-to-end testing. Finally, we commented on the need for SSA to resolve stakeholder concerns to ensure program acceptance and use, and take additional steps to consult with the medical community. SSA acknowledged the importance of ensuring sound relations with stakeholders and the need to take additional actions, where necessary, to ensure that all stakeholder concerns have been adequately addressed. SSA also stated that additional steps would be taken to keep stakeholders involved in the initiative, and that plans were being made for a meeting with state DDS representatives to discuss the AeDib initiative. However, SSA has not yet provided us with a copy of its communications plan for dealing with stakeholder issues, including its plans for consultation with the medical community. As part of our ongoing work, we will continue to monitor SSA's progress in addressing these issues. 4. The SSA has worked 11 years now to implement an electronic disability folder. The automation project started in 1992 as the Modernized Disability System (later renamed the Reengineered Disability System). After this project failed in 1999, the SSA immediately began work on the accelerated electronic disability initiative. Why is this project so difficult for the agency to complete? Has the SSA learned from its mistakes along the way? Software development is one of the riskiest areas of systems development.[Footnote 4] We have reported that SSA's software development efforts have been problematic and plagued with delays because SSA has not consistently followed sound practices in developing systems designed to automate its disability claims process; thus, it has experienced numerous software development problems over the past 11 years. For example, in September 1996 we reported that software development problems had delayed the scheduled implementation of the Reengineered Disability System (RDS) by more than 2 years.[Footnote 5] An assessment of the development activity revealed a number of factors as having contributed to that delay, including (1) using programmers with insufficient experience, (2) using software development tools that did not perform effectively, and (3) establishing initial software development schedules that were too optimistic. We reported again, in June 1998, that SSA had encountered performance problems during its RDS pilot tests.[Footnote 6] In response to these performance problems, SSA delayed its plans for expanding the pilot to other offices, and obtained a contractor to independently evaluate and recommend options for proceeding with the initiative. SSA's contractor reported that RDS software had defects that would diminish the case-processing rate at DDS sites, and that SSA had not been timely in addressing software defects. For example, 90 software problems identified by SSA staff remained unresolved for more than 120 days. As a result, the contractor recommended that SSA discontinue the RDS initiative and focus on an alternative solution involving the use of an electronic folder to replace the paper-based case folder in the disability determination process. In another example, we reported in August 2001 on weaknesses regarding SSA's adherence to key software development procedures for several projects, including its electronic disability system.[Footnote 7] We noted that SSA did not consistently adhere to its software development procedures in the areas of requirements management, software project planning, software quality assurance, and software configuration management. In our ongoing review, we have found that SSA has taken important steps to mitigate past software development problems. It has generally addressed its contractor's recommendations aimed at automating its disability claims process, and has generally been applying key software process improvement practices to its development of AeDib projects, such as developing plans to manage the projects, tracking and overseeing the initiatives to measure progress, performing quality assurance reviews to determine that the project is complying with its policies and procedures, and performing configuration management activities. While continually applying these software development practices is no guarantee of AeDib success, these practices nevertheless should improve SSA's capability to develop high-quality software in support of AeDib, thereby avoiding mistakes experienced in the past. Agency Comments and Our Evaluation: The Commissioner of Social Security provided comments on a draft of this correspondence, which are reproduced in full as the attachment. In her comments, the Commissioner offered clarifications to our responses to questions 1 through 3. Regarding question 1, the Commissioner said that we were incorrect in stating that SSA's cost-benefit analysis had not considered all costs of certain critical IT infrastructure elements supporting the nationwide AeDib rollout, and that, in fact, SSA had included the costs of scanning and imaging by the outside vendor in its estimates. We have revised our response to clarify that SSA's cost-benefit analysis considered some, but not all, of the key cost elements that could affect the initiative. SSA's February 2003 cost-benefit analysis noted, for example, that the agency had not considered as part of its scanning and imaging functions, the keying in of indexing information (for case folder identification) by the outsourced scanning vendor, although this is deemed critical to the implementation of SSA's document management capability. Further, during our review, SSA officials told us that certain costs associated with the scanning and imaging functions were not expected to be identified until the agency performed its pilot tests (now ongoing) for its document management architecture. The Commissioner added that other AeDib-related costs associated with telecommunications, disaster recovery, and on-site retention and destruction of source documents should more appropriately be accounted for as part of the agency's ongoing operations, and therefore were covered in the agency's regular infrastructure costs, rather than the AeDib cost estimates. However, we disagree. OMB guidance states that cost-benefit analyses should include comprehensive estimates of all direct and indirect costs associated with a project.[Footnote 8] As such, a sound cost-benefit analysis for AeDib will depend on SSA's fully considering the project-related costs for these critical elements supporting the development, operation, and maintenance of the electronic disability system. Further, the Commissioner expressed concern about the example that we provided in noting that SSA's estimate of AeDib benefits could be overstated. Based on our interviews with SSA and state DDS representatives, our response highlighted the possibility that AeDib benefits could be overstated because of differences in SSA's and the DDSs' assumptions about the extent that medical evidence may be received electronically. We are continuing to assess SSA's cost-benefit analysis as part of our ongoing review. In commenting on our response to question 2, the Commissioner stated that SSA had expected to receive draft risk assessments by June 30, 2003, and either planned to or already had shared some of its risk assessments with us. We have revised our response to reflect that SSA recently provided us with detailed risk assessments for four of the five AeDib projects. Finally, regarding question 3, the Commissioner stated that SSA had provided us with information concerning stakeholder issues and the agency's plans for consulting with the medical community. She further noted that outreach to the medical community was occurring. We are encouraged that SSA is taking steps to ensure productive communications with its key stakeholders and the medical community, and look forward to reviewing documented evidence of the agency's actions. To date, however, we have not received the additional information on SSA's plans for addressing stakeholder issues or consulting with the medical community, that the Commissioner mentions in her letter. : In responding to these questions, we relied on past work and our ongoing review of SSA's efforts to automate its disability claims process. We discussed our assessment of the cost-benefit analysis with SSA's Office of Disability Programs, Office of Systems, and Office of the Chief Information Officer, and with the supporting contractor and key stakeholders. We reviewed and analyzed the most recent agency documents associated with SSA's risk management efforts. We also discussed with officials in the Office of Disability Programs SSA's efforts to develop a strategy for resolving stakeholder concerns, as well as a more aggressive approach for consultation with the medical community. Finally, we reviewed and analyzed SSA's software process improvement documentation, as well as past assessments of the agency's failed attempts to automate its disability claims process to determine lessons learned, and whether SSA is avoiding past software development problems. We conducted our work in accordance with generally accepted government auditing standards, during August 2003. We are sending copies of this letter to the Honorable Jo Anne B. Barnhart, Commissioner of Social Security, and other interested parties. Copies will also be available at our Web site at [Hyperlink, www.gao.gov.] w [Hyperlink, http://www.gao.gov] ww.gao.gov. If you have any questions on matters discussed in this letter, please contact me at (202) 512-6240 or Valerie Melvin, Assistant Director, at (202) 512-6304. We can also be reached by E-mail at [Hyperlink, email@example.com] k [Hyperlink, firstname.lastname@example.org] email@example.com and [Hyperlink, firstname.lastname@example.org] m [Hyperlink, email@example.com] firstname.lastname@example.org, respectively. Key contributors to this correspondence include Michael A. Alexander, Tonia B. Brown, and Mary J. Dorsey. Sincerely yours, Linda D. Koontz Director, Information Management Issues: Signed by Linda D. Koontz: [End of section] Enclosure: Comments from Social Security: SOCIAL SECURITY: The Commissioner: September 3, 2003: Ms. Linda D. Koontz: Director, Information Management Issues United States General Accounting Office 441 G Street N. W. Washington, D.C 20548: Dear Ms. Koontz: This responds to your letter dated August 27, 2003, requesting our comments on your proposed correspondence entitled Social Security Administration: Subcommittee Questions Concerning Efforts to Automate the Disability Claims Process (GAO-03-1113R). I appreciate the opportunity to review and comment on the correspondence before it is issued. We have the following comments and clarifications on your proposed responses to questions 1 through 3. Question 1: In your draft note to Congressman Shaw you indicate that in the February 2003 cost-benefit analysis (CBA) the Social Security Administration (SSA) "...did not consider all costs of certain critical information technology infrastructure elements supporting the nationwide rollout, such as scanning and imaging by the outsource vendor, telecommunications, disaster recovery and on-site retention and destruction of source documents, such as medical records." This statement is incorrect, since we did in fact include the costs of scanning and imaging by the outsource vendor in the estimated costs for the accelerated electronic disability system (AeDib). Pages 8 and 9 of the CBA (Version 3.1 dated February 3, 2003) list these costs among the significant cost impacts for AeDib, with a pie chart showing outsource scanning costs as 14 percent of the total life cycle costs. We consider the other costs to be part of our ongoing operations, rather than specifically related to AeDib, and these costs are more appropriately covered in our regular infrastructure costs. Your draft response also indicates that the benefits of AeDib may have been overestimated because the State Disability Determination Services (DDS) you contacted disagree with the estimate that 30 percent of all medical evidence will be received in electronic form by 2004. During the fall of 2002, we held several discussions with the DDS Systems Committee and the National Council of Disability Determination Directors leadership relative to business process and technological issues. One of the central topics of these discussions was how to gather information that would assist in building the electronic medical evidence (EME) estimates. We derived the 30 percent estimate based on these discussions and information received from the DDS community. This is a national estimate for fiscal year 2004, and we would expect that this level would vary from location to location. Included in the 30 percent estimate is medical evidence received through both direct electronic submissions as well as through fax submittals. Question 2: As stated in your draft note, we have provided the comprehensive risk assessments for two of the five parts of AeDib. We had advised General Accounting Office (GAO) staff that we were to receive the draft risk assessments from our contractor by June 30, 2003. We then needed to conduct an internal review of these documents. We shared two additional risk assessments with you on August 28, 2003. The last one will be sent after we complete our internal review. Question 3: Our comments on Question 2 also apply to the response to Question 3. Additionally, we recently provided GAO staff with information regarding stakeholder issues and our plans for consultation with the medical community. Our senior executive leadership meets weekly to address AeDib issues, including those raised by all entities, internal as well as external. Additionally, our outreach to the medical community is handled through our public affairs and professional relations personnel in the DDSs and in SSA's central office, Regional Offices, and Field Offices. The issues of EME and processing disability claims in an electronic environment are being institutionalized throughout SSA. They have been apart of, and will continue to be included in, our contacts and communication vehicles that are already in place, including communication plans. Furthermore, I have undertaken special efforts to communicate with the major medical associations by holding and personally chairing a series of meetings to address their issues. In addition, my staff has had a series of productive meetings with staff at the Department of Health and Human Services on these issues. If I may be of further assistance, please do not hesitate to contact me or have your staff contact Mr. Robert M. Wilson, Deputy Commissioner for Legislation and Congressional Affairs, at (202) 358-6030. Sincerely, Jo Anne B. Barnhart: Signed by Jo Anne B. Barnhart: [End of section] (310386): FOOTNOTES  U.S. General Accounting Office, Electronic Disability Claims Processing: Social Security Administration's Accelerated Strategy Faces Significant Risks, GAO-03-984T (Washington, D.C.: July 24, 2003).  U.S. General Accounting Office, Information Technology: Greater Use of Best Practices Can Reduce Risks in Acquiring Defense Health Care System, GAO-02-345 (Washington, D.C.: September 26, 2002).  Social Security Administration, SSA AeDib Risk Management Plan, Version 2.01 (Baltimore, MD, January 21, 2003).  U.S. General Accounting Office, Social Security Administration: Information Technology Challenges Facing the Commissioner, GAO/T-AIMD- 98-109 (Washington, D.C.: March 12, 1998).  U.S. General Accounting Office, Social Security Administration: Effective Leadership Needed to Meet Daunting Challenges, GAO/HEHS-96- 196 (Washington, D.C.: September 12, 1996).  U.S. General Accounting Office, Social Security Administration: Technical and Performance Challenges Threaten Progress of Modernization, GAO/AIMD-98-136 (Washington, D.C.: June 19, 1998).  U.S. General Accounting Office, Information Technology Management: Social Security Administration Practices Can Be Improved, GAO-01-961 (Washington, D.C.: August 21, 2001).  Office of Management and Budget Circular No. A-94 Revised (Transmittal Memo No. 64), Guidelines and Discount Rates for Benefit- Cost Analysis of Federal Programs, October 29, 1992.