MEMORANDUM

The attached evaluation report presents the results of our review. Our objective was to determine the current status of five of the Social Security Administration’s disability process improvement initiatives: Prototype, Quality Assurance, Disability Claims Manager, Process Unification and Hearings Process Improvement.

Please comment within 60 days from the date of this memorandum on corrective action taken or planned on each recommendation. If you wish to discuss the final report, please call me or have your staff contact Steven L. Schaeffer, Assistant Inspector General for Audit, at (410) 965-9700.

James G. Huse, Jr.

Executive Summary

Our objective was to determine the current status of five of the Social Security Administration’s (SSA) disability process improvement initiatives: Prototype, Quality Assurance (QA), Disability Claims Manager (DCM), Process Unification, and Hearings Process Improvement (HPI).

In September 1994, SSA released the Plan for a New Disability Claim Process in response to increasing case workloads and processing times and concerns with reversal rates at the hearings level. Over the next few years, SSA made little progress in implementing the initiatives outlined in the plan. At the advice of the General Accounting Office in 1996, SSA subsequently scaled back the number of initiatives that it would concentrate on completing.

In March 1999, the Commissioner announced several decisions about the initiatives for improving the disability claim process. Among his decisions were to:

• Combine the most promising features from the Full Process Model initiative into a Prototype, and evaluate the combination of features.
• Develop a more comprehensive and uniform QA system that improves the review of disability determinations nationwide.
• Continue testing the DCM position through Fiscal Year (FY) 2000.
• Continue Process Unification efforts.
• Continue hearings process improvements, eliminating numerous hand-offs and inadequate tracking of cases.

SSA has not made as much progress implementing an improved disability determination process as originally envisioned in the Plan for a New Disability Claim Process. The new disability claim process was to be fully implemented by FY 2001, but SSA’s timelines have been frequently revised to accommodate changes in the initiatives.

SSA continues to plan, test, and make decisions on implementing initiatives to improve the disability claims process. Of the five initiatives we reviewed, the HPI and Process Unification initiatives have been implemented nationwide. Of the remaining initiatives (1) the Prototype initiative is on hold pending the analysis of program costs and appeal rates to the Office of Hearings and Appeals (OHA); (2) the plan for the QA initiative has not been fully developed; and (3) the DCM initiative did not show a significant overall improvement in claims processing and was not implemented.

Decisions about the expansion of the Prototype initiative at additional DDSs were delayed. Preliminary data from the Prototype DDSs raised questions about the program costs of national implementation. In addition, the number of Prototype cases that were appealed to OHA was higher than SSAanticipated. SSA has not released information on program costs; however, it has reported that the resources freed by eliminating the reconsideration level will not cover the costs of the Prototype process as originally thought. In addition, the number of Prototype cases that were appealed to OHA was higher than SSA anticipated.

Prototype operations continue in the original 10 DDSs. Prototype results to date are mixed. More allowances are being made earlier in the process, which may result in improved citizen satisfaction. However, productivity at the Prototype DDSs has decreased about 13 percent and processing times have increased about 23 percent. Furthermore, preliminary data indicate that approximately 25 percent of the Prototype DDSs’ cases were appealed to OHA. This appeal rate is 6 percent higher than the appeal rate at the DDSs used for comparison purposes. The appeal rates are not final because baseline differences between the Prototype and the comparison DDSs have not been addressed.

We noted that some Prototype DDSs have experienced higher than average disability examiner (DE) attrition rates. In FYs 2000 and 2001, 6 of the 10 Prototype DDSs (Alabama, California, Colorado, New Hampshire, New York, and Pennsylvania) had higher DE attrition rates as compared to their 1997 rates. Also, 6 of the 10 Prototype DDSs (Alabama, Alaska, Colorado, Louisiana, Missouri, and New Hampshire) had DE attrition rates that exceeded the national average of 13 percent in FY 2000, ranging from 17 to 24 percent. Five of the same 6 Prototype DDSs (except for Alaska) had rates that exceeded the national average of 13 percent in FY 2001, ranging from 14 to 22 percent. Each State’s administration of its DDS can impact attrition rates, e.g., through salary levels. However, the implementation of the Prototype and Process Unification initiatives represents a period of change for DEs. As a result of these initiatives, DEs have experienced a significant increase in responsibilities and required skills. We did not confirm if these changes contributed directly to the attrition rates at the Prototype DDSs.

To date, SSA does not have a fully developed plan for a more comprehensive and uniform QA system. In reviewing SSA’s QA system, a contractor informed SSA that simply modifying the system or adding resources would not move SSA toward its quality improvement goals. Instead, SSA should adopt an advanced quality management system. In July 2001, the Acting Commissioner appointed a senior-level steering committee to develop recommendations for proceeding with the QA initiative. The committee met with the Commissioner to discuss the contractor’s reports and several possible QA pilots. The Commissioner has asked for more details about the QA pilots and the findings of the contractor.

On October 22, 2001, the Acting Commissioner announced that the DCM initiative would not be implemented at this time. The DCM test results showed that case-processing costs increased and more resources would be needed to support a blended Federal/State process.

The Process Unification initiative is a central theme of all disability process improvements and is viewed as a continual process without an end date. Through this initiative SSA seeks to resolve the inconsistencies or sources of error between the DDS and the OHA levels of decision making. SSA is continuing to issue policy statements in the same language to all adjudicators in the disability determination process and to clarify policies contributing to inconsistent decision making by producing regulations and Social Security Rulings that are binding on all levels of adjudication. The Commissioner has not announced any new plans for this initiative.

The HPI initiative has been implemented in all hearings offices, with Phase II and Phase III hearings offices completing the first year of operation in November 2001. HPI was developed to increase OHA productivity through improvements in the internal work processes and automated systems. Since FY 1999, productivity has decreased about 11 percent and processing times have improved only slightly. Furthermore, the planned systems technology has not been completely implemented. In July 2001, SSA announced that a comprehensive review of the HPI initiative would be conducted by a new steering committee. The steering committee presented the results of its review and recommendations to the Commissioner in January 2002. The Commissioner has not announced any decisions on the future of HPI.

Since 1994, SSA has aspired to improve customer service and adjudicate disability cases in a more timely, accurate, and efficient manner by means of its disability improvement initiatives. To date, these initiatives have not resulted in significant improvements in the disability claims process.

Prototype DDSs are falling behind in providing timely customer service as processing time at the initial level has increased, production has decreased, and the backlog of cases is growing at DDSs and OHA. Furthermore, the Agency is concerned with the program costs of implementing the Prototype process nationally and the appeal rates of Prototype cases were higher than SSA anticipated. Nonetheless, expected benefits of the Prototype process include, but are not limited to, improved claims accuracy, increased productivity, decreased claims processing times, and improved citizen satisfaction. SSA should evaluate whether the increased program, DDS, and OHA costs resulting from Prototype operations justify the benefits offered by the new process.

Retaining qualified DEs is essential to the success of Prototype operations. However, some Prototype DDSs have experienced higher than average DE attrition rates. SSA must determine the reasons for the Prototype DE attrition before decisions are made on the national implementation of Prototype.

A uniform, consistent and accurate QA system is critical to disability claim adjudication. SSA needs to be proactive in developing and implementing a new QA system.

SSA’s evaluation of DCM concluded that the costs of implementing DCM would not justify the modest improvements in performance. While the results from the DCM initiative revealed higher costs, there were nonmonetary benefits to DCM, such as improved customer satisfaction. We did not determine if the nonmonetary benefits justified the higher costs of DCM. Accordingly, we did not reach an overall conclusion regarding SSA’s decision not to implement DCM.

The overall theme of Process Unification is to resolve decisional inconsistencies between the DDS and OHA levels of decision making. We were unable to determine any direct affects of Process Unification on the disability claims process. Until complete data on OHA allowance rates for Prototype cases become available, it is impossible to independently assess whether Process Unification has resulted in any measurable improvements to the disability process.

HPI has not resulted in significant improvements in processing times and has resulted in decreased productivity. However, these results may improve as OHA staff becomes more familiar with HPI and as more enhanced automation support is implemented. Implementation of improved automation at OHA could help to decrease case processing times and improve management's monitoring of case processing.

We recommend that SSA:

• Proceed with national implementation of Prototype only if the benefits of the process justify the increased program, DDS, and OHA costs.

• Evaluate DE attrition at the 10 Prototype DDSs and take appropriate steps to reduce the rates. This evaluation should be completed before decisions are made on the national implementation of Prototype.

• Develop and implement a comprehensive QA system that produces accurate and uniform disability determinations nationwide. In doing so, SSA should establish a timeline for developing and implementing the new QA system and monitor completion of the key milestones.

• Assess the impact of Process Unification when data on OHA allowance rates for Prototype claims are available.

• Implement the enhanced automation at OHA as outlined in the HPI plan if this initiative continues.The automation may provide management with the information necessary to improve case processing, development and scheduling.

In response to our draft report, SSA agreed with all of our recommendations. SSA also outlined decisions made on the disability process improvement initiatives after our draft report was issued. General and technical comments were provided, and we incorporated them into the report as appropriate. See Appendix G for the full text of SSA's comments to our draft report.

Table of Contents

Page

Appendix B – Prototype

Appendix C – Quality Assurance

Appendix D – Disability Claims Manager

Appendix E – Process Unification

Appendix F – Hearings Process Improvement

Appendix G – Agency Comments

Appendix H – OIG Contacts and Staff Acknowledgments

AC Appeals Council
ALJ Administrative Law Judge
DCM Disability Claims Manager
DDS Disability Determination Service
DE Disability Examiner
DQB Disability Quality Branch
FPM Full Process Model
FY Fiscal Year
GAO United States General Accounting Office
HO Hearing Office
HOD Hearing Office Director
HPI Hearings Process Improvement
Lewin The Lewin Group, Inc.
MC Medical Consultant
MI&E Management Information and Evaluation Workgroup
OHA Office of Hearings and Appeals
OQA Office of Quality Assurance and Performance Assessment
PEM Pugh Ettinger McCarthy Associates, L.L.C.
PHC Pre-Hearing Conference
POMS Program Operations Manual System
PPWY Production Per Work Year
QA Quality Assurance
RFC Residual Functional Capacity
RRP Regional Review Panel
SDM Single Decision Maker
SSA Social Security Administration
SSAB Social Security Advisory Board
SSR Social Security Ruling

Introduction

Our objective was to determine the current status of five of the Social Security Administration’s (SSA) disability process improvement initiatives: Prototype, Quality Assurance (QA), Disability Claims Manager (DCM), Process Unification, and Hearings Process Improvement (HPI).

In September 1994, SSA released the plan for an improved disability claim process in response to increased Disability Determination Services (DDS) caseloads and processing times, and concerns with reversal rates at the Office of Hearings and Appeals (OHA). The plan included five primary objectives:

• the process is user-friendly for claimants and employees;

• an allowance decision, if applicable, is made as early in the process as possible;

• all disability decisions are made and effectuated quickly;

• the process is efficient; and

• employees find the work satisfying.

In the 1994 plan, SSA proposed an ambitious series of initiatives to improve timeliness, accuracy and customer service. SSA committed to 83 initiatives to be accomplished over 6 years. When the plan was issued, SSA estimated that improving the disability process would cost $148 million and yield net savings of $704 million through Fiscal Year (FY) 2001. Once the 1994 improvement plan was fully implemented in FY 2001, annual savings of $305 million would result.

In 1996, the General Accounting Office (GAO) concluded that SSA’s plan was overly ambitious and complex. At that time, SSA had made little progress in meeting its goals, could not demonstrate positive results, and faced difficulty retaining the support of some stakeholders. In response to the urging of GAO and stakeholders, SSA issued a scaled-back disability process improvement plan in February 1997. The revised plan highlighted eight key initiatives to be accomplished over 9 years: Single Decision Maker (SDM), Adjudication Officer, Full Process Model (FPM), Process Unification, QA, Simplified Decision Methodology, Redesigned Disability System, and DCM.

After 2 years of testing initiatives, the Commissioner announced several decisions about the improvement efforts in March 1999. Among his decisions were to:

• Combine the most promising features from the FPM into a Prototype, and evaluate the combination of features.
• Continue to develop a more comprehensive and uniform QA system that improves the review of disability determinations nationwide.
• Continue testing the DCM position through FY 2000.
• Continue Process Unification.
• Continue with hearings process improvements to eliminate numerous hand-offs and inadequate tracking of cases.

See Appendix A for details on the history of SSA’s disability process improvement initiatives.

To determine the status of the five initiatives, we reviewed each initiative’s historical background (see Appendix A) and each initiative’s progress from March 1999 to October 2001. We also determined SSA’s future plans for the initiatives.

To accomplish our objective, we:

• reviewed applicable sections of Federal Regulations, Social Security Rulings, and SSA's Program Operations Manual System (POMS);

• reviewed available SSA documents on the disability process improvement initiatives;

• reviewed prior audit and evaluation reports related to disability process improvement issued by GAO and the Social Security Advisory Board (SSAB);

• conducted interviews with employees at SSA Headquarters in Baltimore, Maryland, and OHA in Falls Church, Virginia;

• reviewed FY 2001 Congressional testimony about the initiatives; and

• reviewed performance data from employees at OHA.

We did not verify the accuracy or validity of the SSA data used in this report.

We performed fieldwork in Baltimore, Maryland; Falls Church, Virginia; and Kansas City, Missouri; from February to October 2001.

We conducted our evaluation in accordance with the Quality Standards for Inspections issued by the President's Council on Integrity and Efficiency.

Results of Review

SSA has not made as much progress implementing an improved disability determination process as originally envisioned in the Plan for a New Disability Claim Process. The new disability claim process was to be fully implemented by FY 2001. However, SSA’s timelines have been extended to accommodate changes in the initiatives.

SSA continues to plan, test, and make decisions on implementing initiatives to improve the disability claims process. Of the five initiatives we reviewed, the HPI and Process Unification initiatives have been implemented nationwide. Of the remaining initiatives (1) the Prototype initiative is on hold pending the analysis of program costs and appeal rates to Office of Hearings and Appeals; (2) the plan for the QA initiative has not been fully developed; and (3) the DCM initiative did not show a significant overall improvement in claims processing and was not implemented.

• SDM and a revised role for the medical consultant to give disability examiners (DE) greater decisional authority while more effectively using medical consultants’ expertise;

• claimant conference to allow claimants facing a denial decision another opportunity to provide additional evidence;

• enhanced documentation and explanation (rationale) to require more complete case development and explanation of the disability determination;

• removal of the reconsideration level from the current four-step process to eliminate an average of 70 days processing time and make those resources available for use at the initial level.

National rollout of the Prototype initiative was scheduled to begin in April 2002. However, as of May 2001, implementation of the initiative was delayed pending analysis of program costs of national implementation. Also, SSA is concerned with the appeal rates of Prototype cases because of the impact on OHA’s workload to include extended processing times. The Prototype will continue to operate in the same 10 DDSs, but the initiative will not be implemented in any additional DDSs at this time.

The Prototype interim report was released by SSA in July 2001. This report and other information sources disclosed the following results of the Prototype initiative testing:

• The initial disability determination process became more efficient by eliminating the reconsideration level (one less administrative step and about 70 days less processing time) and some unnecessary reviews by medical consultants (see Appendix B, page B-4 for details).

• Claimants with a less-than-fully-favorable decision are offered a claimant conference with the decision maker. Claimant conference participation has increased from 56 percent in the FPM to 64 percent in the Prototype. Participants, especially those who were awarded disability benefits, rated satisfaction as fairly high (see Appendix B, page B-6 for details).

• Compared to FY 1999, allowances were 1.5 percent higher and claimants allowed at the initial level in the Prototype process were allowed 135 days earlier on the average than those going through the current reconsideration process (see Appendix B, page B-7 for details).

• Initial processing times increased 23 percent from FY 1999 to FY 2001 because claimant conferences and enhanced documentation and explanation (rationale) took more processing time. Since FY 1999, productivity decreased by 13.4 percent (from 253 case clearances to 219 clearances per workyear) and backlogs grew (see Appendix B, pages B-7 to B-9). According to the Office of Disability, lower productivity (based on production-per-workyear statistics) and higher processing times (due in part to the claimant conference and enhanced documentation and explanation) were expected.

• Additional costs of the new process exceeded the resources saved by eliminating the reconsideration level (see Appendix B, pages B-9 to B-11).

• Accuracy rates did not significantly change. Denial accuracy remained a problem for Prototype and nonPrototype States (see Appendix B, page B-4).

• The appeal rates to OHA for Prototype cases are higher than SSA expected. The Office of Quality Assurance and Performance Assessment (OQA) is tracking about 200,000 cases from the DDS level through the appeals level (see Appendix B, page B-14). OQA informed us that approximately 25 percent of initial claims in Prototype DDSs were appealed to OHA. This is a significant increase when considering that 19 percent of initial claims were appealed to OHA in the States used for comparison, which are operating under the current process. However, the appeal rates are not final because baseline differences between the Prototype and the comparison DDSs have not been addressed. Baseline data must be analyzed to ensure that known differences for the Prototype and comparison groups are properly considered.

National implementation of the Prototype initiative is delayed pending analysis of program costs and concerns about appeal rates on public service and overall processing times. SSA estimates that analysis of cost impacts will be complete in spring 2002, including a full evaluation of the impact of the Prototype initiative on allowance rates. SSA created the Prototype Process Refinements Team to make adjustments to policies and procedures, revise regulatory language, and make other refinements to the disability process. See Appendix B for further details.

As a result of the Prototype and Process Unification initiatives, DEs have experienced a significant increase in their responsibilities and required skills. Now required are higher level analytic thinking and writing skills, more verbal skills, interviewing skills, and skills for dealing with the public. The DE must have both adequate knowledge about the medical and vocational aspects of disability and the ability to apply Process Unification rulings in the adjudication process. The determination process features more emphasis on claimant statements of pain and other symptoms, determining the credibility of the statements, determining the claimant’s functional limitations, and resolving inconsistencies in medical evidence.

Because the DE position has become more complex, concerns have been voiced that not all current DEs can become SDMs. The Philadelphia workgroups in September 2000 recommended that SSA begin discussions about effectively using adjudicators who are unable to become SDMs. Monitoring visits conducted by the SSAB in California also revealed that a number of DEs are unable to acquire the skills and apply the knowledge necessary to process disability cases according to Process Unification principles. In addition, the SSAB found that examiner attrition and hiring qualified employees to replace DEs who leave are problems.

The inability of some current DEs to become SDMs and the pre-existing program-wide examiner attrition problem highlight the deepening crisis of recruiting and retaining qualified DEs. Examiner attrition is a long-standing problem for DDSs nationwide, and the problem has worsened in the Prototype DDSs. Anecdotally, the Office of Disability told us some reasons why examiners might leave employment at DDSs. However, except for maintaining attrition rates by States, the Office of Disability is not systematically collecting evidence from DEs to assist in solving the attrition problem.

Possible ways to increase the pool of qualified applicants and retain current examiners are revising and updating the position description and training regimen for DEs, to upgrade the examiner position by instituting professional certification, and increasing salary and benefits. SSA has been proactive in some ways in preparing new examiners and retaining existing DEs. In September 2000, SSA released a training package for new DEs to provide the foundation for learning the disability adjudication process. SSA has provided refresher training for DEs on a number of topics related to the adjudication process, and more training courses are forthcoming. SSA is considering the feasibility of examiner certification, which could ensure a standard level of proficiency and performance and give recognition of professional status. The Disability Training Steering Committee has identified basic knowledge, skills and abilities for the examiner position and asked for input from DDSs regarding knowledge, skills, and training for examiners. This information might be used to help State governments review the DE position for possible upgrades.

In regard to raising salaries for disability examiners, the Prototype States of Michigan and New York ranked in the top five States nationally for examiner compensation in FY 1999. Average yearly compensation was approximately $55,000 for Michigan and $61,000 for New York, as compared to the national average of about $45,000. Along with their above average compensation, Michigan and New York had examiner attrition rates of less than five percent in FY 2000, although in FY 2001 both States experienced an increase in the attrition rate, as shown in Chart 2,Chart 2, Appendix B. Barriers to increasing the compensation of examiners might be the increased cost to SSA and the State government’s control over setting salary levels.

Because DEs are the linchpin of the disability determination process, the success or failure of disability process improvement initiatives depends on recruiting and retaining qualified DEs. Examiner attrition is a serious problem and needs to be continually addressed by SSA in a systematic and effective way.

OQA is responsible for performing quality assurance reviews on the accuracy of disability decisions at the DDS and OHA levels. Current quality assurance reviews are end-of-line reviews; i.e., the accuracy of the disability decision is reviewed rather than examining the in-line quality or accuracy of the whole decision-making process. At the DDS level, for example, the Disability Quality Branch (DQB), located in the SSA Regional offices, reviews DDS determinations through two end-of-line mechanisms, the quality assurance review and the pre-effectuation review. If deficiencies or inaccuracies are found, the DQB returns cases to the DDS for correction. Another end-of-line mechanism, the consistency review, is performed by OQA in Headquarters to determine if DQBs are consistently applying SSA’s quality review standards and achieving equitable treatment of claimants across all States. OQA reports on the accuracy of decisions by calculating the DDS Decisional Accuracy indicator and the OHA Decisional Accuracy indicator.

The objective of the QA initiative is to provide for accurate, uniform and consistent disability adjudication nationwide. SSA contracted with The Lewin Group, Inc. (Lewin) and Pugh Ettinger McCarthy Associates, L.L.C. (PEM Associates) in December 1999 to assess the present QA system and identify what SSA could do to build an effective and comprehensive QA system.

The final report was delivered to SSA in March 2001 and included short-term and long-term options in eight areas. In brief, the contractor concluded that modifying the system or adding resources would not move SSA toward its quality improvement goals and end its reliance on end-of-line review. Instead, the contractor stated that SSA should adopt an Agency-wide advanced quality management approach. See Appendix C for more details on the QA initiative.

The only new activity related to the QA initiative is an OQA pilot study of a new consistency review process. The pilot was implemented in February 2001 and will continue for a year. The aim of consistency reviews is to locate and resolve inconsistencies in the application of policy and adjudication at the State (DDS) and Federal (Regional and Headquarters) levels. Under the old process, DQBs review DDS determinations then send a subsample of decisions with errors to Headquarters for review. The improved consistency review involves sending a sample of DDS determinations to both the DQB and Headquarters; any inconsistencies in decisions are resolved at the Federal level and shared with the DDSs. Also, a case with a specific issue is sent to both the DQBs and DDSs for their determination. The determinations are sent to a multi-component panel for analysis, and an appropriate solution is returned to the DQBs and DDSs.

The future of the QA initiative depends on the Agency’s decision about the type, scope, and costs of a new QA system. The Acting Commissioner appointed a senior-level steering committee in July 2001 to consider Lewin and PEM Associates’ recommendations, as well as input from other sources. He charged the committee to develop recommendations for a new quality process and a new quality culture within the Agency. The work of the committee was to become a part of SSA’s Integrated Disability Plan and used by the new Commissioner as a tool for decision making. The committee met with the Commissioner in early FY 2002 to discuss the Lewin and PEM Associates’ reports and several possible QA pilots. The Commissioner has requested more details about the reports and the QA pilots.

The DCM position was created to combine the title II and title XVI claims representative duties in field offices with the DDS’ DE duties. The DCM was to serve as the single point of contact for the claimant, managing the disability claim from intake through the initial disability determination, including all medical and non-medical claims activities. Originally, the concept was to support the DCM with a "simplified decision methodology" and automated improvements. These supports did not materialize. In 1997, SSA decided to proceed with testing the DCM position without the supports.

DCM was tested at 36 sites in 15 States and was located in both field offices (at the Federal level) and DDSs (at the State level). The DCM handled only adult title II and title XVI disability claims; no title XVI children’s cases were included. Both State and Federal employees were trained to function as DCMs.

Phase I, training and on-the-job training was implemented in November 1997 and continued through June 1999. Lewin was contracted to help SSA assess Phase I and to develop recommendations for conducting the Phase II evaluation of the DCM test. According to Lewin, the Phase I assessment indicated the DCM was "a viable approach" to claims processing, showing comparable accuracy, somewhat faster processing time, lower productivity (which can be improved), and higher allowance rates than the current process.

Phase II, the formal evaluation, was performed from November 1999 through November 2000 to ensure the requisite number of control cases was processed for a statistically valid evaluation. Lewin evaluated the draft report on several criteria before SSA released it in June 2001.

In January 2001, the Commissioner stated that concerns emerged about the cost of the process and the long-term viability of maintaining skill levels of the position. Also, DCM implementation would pose key challenges in the Federal-State relationship.

The final report was issued October 19, 2001. Findings of the DCM report were:

• DCM overall processing time was faster than the control group.

• DCM initial allowance rate and the cumulative allowance rate through the reconsideration level was about the same as compared to the control group.

• DCM allowance and denial accuracy rates were about the same as the control group’s accuracy rates.

• DCM productivity ranged from 14 percent less to 8 percent more than the current process. Since some DCMs were located at DDSs and other DCMs were at field offices, two models of productivity and costs were developed so that DDS (State) and field office (Federal) data could be combined and compared to the current process.

• DCM initial claim cost was 7 percent to 21 percent higher than the current process.

• Customer satisfaction of denied claimants was higher as compared to a group surveyed by the FY 2000 Market Measurement Program Survey of Initial Disability applicants.

• DCMs had higher job satisfaction than they did in their prior job.
  
  
• DCM attrition rate was higher compared to the control group’s attrition rate.
  
  
• DCMs in Prototype settings had comparable processing times, allowance rates and accuracy rates as the control group.
  
  
• DCM model, as tested, must be legislated. For more details on each of the findings, see Appendix D.

On October 22, 2001, the Acting Commissioner announced that the DCM initiative would not be implemented at this time. The results of 3 years of testing showed that the costs of implementing the DCM might not justify the modest improvements in processing times and citizen satisfaction.

Process Unification is considered a central theme of all disability process improvement efforts and is viewed as a continuous improvement process without a clear end point. Through this initiative SSA seeks to resolve the inconsistencies or sources of error between the DDS and OHA levels of decision making. The objective of Process Unification is to achieve correct and accurate results on similar cases at all levels of adjudication by means of the consistent application of laws, regulations and rulings.

In summary, the Process Unification features are:

• Single presentation of policy

• Social Security Rulings (SSR)

• Other sub-initiatives addressing DDS and OHA decision-making

• Complete documentation and detailed explanation of decisions (rationale);
• Remand selected hearing cases; and
• Quality review of hearing decisions.

A meeting to discuss issues and to develop recommendations regarding Process Unification principles and Prototype operations was held in Philadelphia in September 2000. Workgroups proposed recommendations focusing on policy, workload, training and applying the SSRs in an operational setting. A status report on the implementation of the Philadelphia workgroups’ recommendations was issued on June 25, 2001. Additional Prototype and Process Unification-related recommendations have been proposed by other SSA groups such as the Process Redesign Refinement Team and the 30-Day Workgroup.

Since the affects of other changes in disability operations are difficult to control for and Process Unification cannot be measured directly, we have to rely on such indirect indicators as the DDS and OHA allowance rates and decisional quality to determine if Process Unification is succeeding. For example, if Process Unification is working as intended, we would expect to see fewer OHA allowances after DDS disability determinations from the Prototype DDSs and a lower OHA allowance rate. While the total OHA workload allowance rate of 58.4 percent for FY 2001 shows a decline from 67.1 percent in FY 1992, we will have more complete information when the data on OHA allowance rates for Prototype claims become available for assessment.

Until the new Commissioner discloses other plans for the Process Unification initiative, SSA will continue to issue policy statements in the same language to all adjudicators in the disability determination process, and SSA will address differences in policy instructions issued before July 1995. SSA will continue to clarify policies contributing to inconsistent decision making by issuing regulations and SSRs that are binding on all levels of adjudication. In a July 2001 memorandum announcing the creation of an integrated plan for disability, the Acting Commissioner stated that further discussion within SSA and by stakeholders is needed to decide whether to pursue specific regulations (or even legislative changes) in addressing some of the most difficult policy issues: (1) the connection between symptoms (such as pain) and an objective medical basis expected to cause the symptoms; and (2) the weight and explanation given to medical source opinions. While these issues have been raised within the Agency for discussion, the new Commissioner has not announced any decisions on Process Unification at the present time.

The Office of Hearings and Appeals (OHA)OHA provides claimants who are denied benefits at the initial and reconsideration levels the opportunity for a full due-process hearing on their claim. The current appeals process has three administrative levels of review. First, the claimant can request that SSA reconsider the initial determination made at the DDS. Second, if the claimant is dissatisfied with the DDS determination at the reconsideration level, the claimant may request a hearing before an Administrative Law Judge (ALJ) in OHA. Third, the claimant may then appeal the ALJ’s decision to the OHA Appeals Council (AC) if the claimant is dissatisfied with the decision. The AC may deny, dismiss or grant the request for review. If the AC grants the request for review, the AC either issues a decision or remands the case to an ALJ. The final recourse for the claimant is the appeal to a Federal district court.

SSA’s concern with the increasing numbers of disability claims and lengthy processing times at the DDS level in the early to mid-1990s was also a concern at the OHA level. From FY 1992 to FY 1996 average processing time for the total OHA workload increased 69.5 percent from 223 days to 378 days.

The need to develop a more effective hearings process from the request to final disposition resulted in the Hearings Process Improvement (HPI) initiative. In FY 1999, the HPI Plan was released by SSA and the following goals were endorsed:

• reduce overall case processing time;

• increase productivity or the number of hearings cases processed (productivity per workyear or PPWY);

• improve the quality of customer service to the claimant;

• promote case management; and

• increase employee job satisfaction.

The HPI initiative is expected to show significant improvements in case processing times and customer service. To achieve these goals, a main feature of the HPI plan is to create a new internal work process for appealed disability cases. Changes to staff functions and responsibilities mean the employee learning curve becomes a factor in processing times. OHA’s pre-HPI and HPI processes are described in the following table.

HPI was implemented in three phases. The first phase began in January 2000 at 37 hearings offices (HO), although May 2000 was the first complete month that all offices were operating under the new process. HPI Phase II and Phase III implementation began in October 2000 at the remaining 101 HOs. December 2000 was the first complete month that all offices were operating under the new process.

OHA statistics in table 2 show little or no improvement in case processing times and case production (PPWY). Average processing time has decreased slightly from 314 days in FY 1999 (pre-HPI) to 308 days in FY 2001, and PPWY indicates an 11 percent decrease in productivity during the same time period. Also in table 2 are revised performance plan targets for FY 2002. Revisions to the FY 2002 processing time target (from 259 days to 330 days) and to the productivity target (111 to 91) reflect congressional action on SSA’s FY 2002 budget (i.e., FY 2002 workload estimates for hearings dispositions and pendings, the impact of HPI, and the available number of ALJs).

According to OHA, factors responsible for high processing times for all types of cases are the moratorium on hiring additional ALJs because of litigation, the length of the learning curve experienced by employees as they adjust to their new responsibilities, and the expected but not-yet-developed automation support.

As specified in the HPI Plan, the success of HPI relies heavily on enhanced automation, electronic data collection and analysis. Automation support is essential for monitoring workflow, tracking case processing and development, transferring case information, scheduling, and managing information reports. This degree of management information support is not in place.

Current data on the fifth goal of HPI—increased employee job satisfaction—are not available yet. However, some preliminary information on employee job satisfaction was collected in August 2000 by the Office of the Inspector General, Office of Audit, and reported in an evaluation of the HPI Phase I implementation. A questionnaire was sent to all employees at Phase I HOs. The responses indicated that (1) employees were continuing to adjust to the changes in the work structure and process instituted by the HPI initiative and (2) differences in perception between managers and nonmanagers were apparent on many questions. Managers (43 percent) and nonmanagers (73.3 percent) viewed the job satisfaction in their HO as worse than before Phase I roll out. Based on categories of high, moderate and low, about 42 percent of managers and 10.8 percent of nonmanagers stated their morale was high.

On July 26, 2001, the Acting Commissioner announced that a comprehensive review of HPI would be conducted by a newly formed steering committee. The committee was charged with identifying improvements to ensure the HPI goals are achieved. An evaluation team analyzed case processing data and conducted site visits at various hearings offices around the country. The steering committee completed its review and briefed the Commissioner on its recommendations in January 2002. The Commissioner has not announced any decisions on the future of HPI.

Conclusions and Recommendations

Since 1994, SSA has aspired to improve customer service and adjudicate disability cases in a more timely, accurate, and efficient manner by means of its disability improvement initiatives. To date, these initiatives have not resulted in significant improvements in the disability claims process.

Prototype DDSs are falling behind in providing timely customer service as processing time at the initial level has increased, production has decreased, and the backlog of cases is growing for the DDSs and for OHA. Furthermore, the Agency is concerned with the program costs of implementing the Prototype process nationally and the number of Prototype cases appealed to OHA exceeded SSA’s expectations. Nonetheless, expected benefits of the Prototype process include, but are not limited to, improved claims accuracy, increased productivity, decreased claims processing times, and improved citizen satisfaction. SSA should evaluate whether the increased program costs and the increased number of appealed cases to OHA resulting from Prototype operations justify the benefits offered by the new process.

Retaining qualified DEs is essential to the success of Prototype operations. However, some Prototype DDSs have experienced higher than average DE attrition rates. SSA must determine the reasons for the Prototype DE attrition before decisions are made on the national implementation of Prototype.

A uniform, consistent and accurate QA system is critical to disability claim adjudication. To date, SSA does not have a fully developed plan for a more comprehensive and uniform QA system. SSA needs to be proactive in developing and implementing a new QA system.

SSA’s evaluation of DCM concluded that the costs of implementing DCM would not justify the modest improvements in performance. While the results from the DCM initiative revealed higher costs, there were non-monetary benefits to DCM, such as improved customer satisfaction. We did not determine if the nonmonetary benefits justified the higher costs of DCM. Accordingly, we did not reach an overall conclusion regarding SSA’s decision not to pursue implementation of the DCM at this time.

The overall theme of Process Unification was to resolve decisional inconsistencies at the DDS and OHA levels. At this time it is impossible to independently assess whether Process Unification has resulted in any measurable improvements to the disability process. We were unable to determine any direct affects that Process Unification may have on the disability claims process. Once data are available on OHA allowance rates for Prototype claims, further assessment of the affects of Process Unification can be made.

HPI has not resulted in significant improvements in processing times and has resulted in decreased productivity. Implementation of improved automation at OHA could help to decrease case processing times and improve management's monitoring of case processing.

We recommend that SSA:

1. Proceed with national implementation of Prototype only if the benefits of the process justify the increased program, DDS, and OHA costs.
2. Evaluate DE attrition at the 10 Prototype DDSs and take appropriate steps to reduce the rates. This evaluation should be completed before decisions are made on the national implementation of Prototype.
3. Develop and implement a comprehensive QA system that produces accurate and uniform disability determinations nationwide. In doing so, SSA should establish a timeline for developing and implementing the new QA system and monitor completion of the key milestones.
4. Assess the impact of Process Unification when data on OHA allowance rates for Prototype claims are available.
5. Implement the enhanced automation at OHA as outlined in the HPI plan if this initiative continues. The automation may provide management with the information necessary to improve case processing, development and scheduling.

In response to our draft report, SSA agreed with all of our recommendations. SSA also outlined decisions made on the disability process improvement initiatives after our draft report was issued. General and technical comments were provided, and we incorporated them into the report as appropriate. See Appendix G for the full text of SSA's comments to our draft report.

Appendices

Appendix B – Prototype

Appendix C – Quality Assurance

Appendix D – Disability Claims Manager

Appendix E – Process Unification

Appendix F – Hearings Process Improvement

Appendix G – Agency Comments

Appendix H – OIG Contacts and Staff Acknowledgments

Appendix A

In September 1994, SSA released the plan for an improved disability claim process. SSA committed to five primary objectives:

• the process is user-friendly for claimants and employees;

• an allowance decision is made as early in the process as possible;

• all disability decisions are made and effectuated quickly;

• the process is efficient; and

• employees find the work satisfying.

SSA’s plan called for 83 initiatives to be accomplished over 6 years with 38 of the initiatives to be implemented within the first 2 years. Because of the complexity of the disability program and the numerous initiatives to be accomplished, SSA made little progress implementing the 38 initiatives. At the advice of the General Accounting Office (GAO) in 1996 and because of stakeholder concerns, SSA subsequently reduced the scope of the disability improvement initiatives and focused on those initiatives SSA considered to be the most important in improving the disability process.

In February 1997, SSA issued a revised plan that endorsed the eight initiatives listed below. The first five initiatives had deadlines before the end of Fiscal Year (FY) 1998, but SSA again made limited progress. SSA’s strategy for testing and implementing many proposed changes at the same time, and problems with the test designs, resulted in delays and unsatisfactory outcomes.

• Single Decision Maker (SDM)—a new position responsible for making the initial determination and consulting with DDS physicians on an as-needed basis.

• Adjudication Officer—a new position to help claimants understand the hearings process, obtain new evidence, request consultative exams, develop cases for the Administrative Law Judge (ALJ) hearing, and issue favorable decisions for clear-cut cases.

• Full Process Model (FPM)—a test of the interactive effects of five features: SDM, Adjudication Officer, use of a predecision interview, elimination of the reconsideration step, and elimination of the Appeals Council (AC) review.

• Process Unification—a series of ongoing initiatives with the objective of achieving similar results on similar cases at all adjudicative levels through consistent applications of laws, regulations, and rulings.

• Quality Assurance (QA)—creation of new in-line processes to build quality into the disability claims review process, and new end-of-line QA to serve as a final review mechanism using one quality standard to review the entire process.

• Simplified Decision Methodology—creation of a timely, efficient, and standardized method for determining who is disabled, focusing on the functional consequences of a person’s medically determinable impairment(s).

• Redesigned Disability System—computer systems support (hardware and software) for improved disability processes.

• Disability Claims Manager (DCM)—a new position blending the field office worker and SDM responsibilities. The DCM is the single point-of-contact for applicants and is responsible for adjudicating claims.

In March 1999, the Commissioner announced several decisions about the disability process improvement efforts. Among his decisions were to:

• Continue testing the DCM position through FY 2000.

• Continue the FPM test to learn more about the effect of eliminating the AC level.

• Continue Process Unification, working toward the goal of similar results on similar cases at all stages of the process through consistent application of laws, regulations and rulings.

• Combine the most promising features from the results of the FPM test into a Prototype, and evaluate the combination of features.

• Eliminate the Adjudication Officer position.

• Establish flexible disability units to improve the capacity to shift disability workloads when necessary to processing centers.

• Continue with hearings process improvements, eliminating numerous hand-offs and inadequate tracking of cases.
  
  
• Continue to develop a fully automated disability system to support an electronic folder (known as e-Dib), which would transmit data from one location to another, including OHA offices.
  
  
• Continue to develop a more comprehensive and uniform quality assurance system to improve the review of disability adjudication across the country.

From our discussions with the Office of Disability, we determined that the goals are to:

• make the disability determination process more efficient and customer friendly;

• realize more allowances earlier in the process;

• ensure the right decision is made the first time;

• appeal fewer decisions to OHA;

• use the same amount of resources allocated to the current process; and

• improve employee satisfaction with the work.

The Management Information and Evaluation (MI&E) Workgroup prepared the Prototype Evaluation Plan and is responsible for collecting, analyzing and reporting information to management throughout the operation of the Prototype. The evaluation’s objectives are to produce information for: (1) adjustments to the Prototype process, (2) national rollout in FY 2002, and (3) the requirements of budget and regulation. At the heart of the evaluation is an impact assessment of program costs, administrative costs, and customer service.

An outside contractor, The Lewin Group, Inc. (Lewin), advised the MI&E Workgroup during the planning phase of the evaluation plan and commented on the efficacy of the plan. Lewin is also responsible for objectively assessing the results of the final evaluation report, which was projected to be completed in December 2001.

SSA monitors each State in the Prototype by collecting and posting weekly data on a SSA website and by conducting site visits. A variety of information on workloads, productivity, allowance rates, processing time, MC involvement and claimant conference participation can be found on the website. It also displays data from the longitudinal database. Site visits involving interviewing DDS staff and administrators were also used to collect more qualitative data about experiences and perceptions during the operation of the Prototype.

The quantitative methodology involves longitudinally tracking approximately 200,000 cases. Disability claims were selected between January 1, 2000 and March 31, 2000 for the Prototype States and between December 1, 1999 and February 29, 2000 for comparison States (one month earlier to allow more time for reconsideration in the current process). The Disability Prototype Longitudinal Database cases will provide information to compare Prototype and comparison States on important indicators such as allowance rates, processing time, appeal rates and accuracy.

Since February 2000 when we began this review, SSA officials have declined our requests for Prototype data, comparison State data, and nonPrototype data for FYs 1999, 2000, and 2001 through March 2001. We were told to wait for the Prototype interim report to be issued in late June 2001. The aggregate data we requested involve costs, processing times, rates and age of pending cases, production, accuracy rates, allowance rates, participation rates in claimant conference and appeals rates.

SSA officials said it was not fair to compare Prototype States with nonPrototype States because States are very different from one another in administration, workload mix, economic and demographic variables, and other factors. If these factors are not adjusted or if known variation is not controlled for, the data comparing Prototype States to nonPrototype States will mislead us and result in incorrect conclusions.

Further, SSA officials advised us not to use data posted at SSA’s Prototype website because this information is regarded as unofficial Agency information and has not been adjusted for known differences between Prototype States and comparison States. It is for weekly monitoring only, and for the discovery of general trends.

According to SSA, comparing Prototype States with comparison States is the only fair and accurate way to determine the progress that the Prototype States are making. We were told that members of the MI&E Workgroup are identifying and quantifying various differences that existed between States before the Prototype implementation, and adjusting data for known differences between each Prototype State and its comparison State. This process will continue until the longitudinal data set is complete; that is, when cases have proceeded through all adjudicative levels.

SSA released the draft interim report to us on August 8, 2001. In this report, SSA estimates that analysis of cost impacts will be complete in Spring 2002. Preliminary results on allowance rates will be available at the end of 2001; a full evaluation, including the projection of the impact of the Prototype on allowance rates and final decisions, will be available in mid-2002.

To report on the status of the Prototype, we have used information from the Prototype interim report, a variety of SSA information sources, and external sources. When the Prototype interim report failed to have the information we needed, we used the latter two information sources.

Streamlining the disability determination process means fewer hand-offs of cases and fewer administrative steps. Efficiency is associated with reduced time and costs. In the Prototype, streamlining and efficiency were accomplished by eliminating the reconsideration level of decision making. By itself, this Prototype feature immediately reduced the number of administrative steps and reduced the case processing time by 70 days.

The purpose of the SDM role is to allow the disability examiner greater decisional independence in the initial determination of disability, which reshapes the role of the MC. The MC provides information and advice to the SDM without the MC being required to sign-off on disability forms (except for claims legislatively mandated; i.e., childhood cases and mental impairment denial cases).

In the histogram below (Chart 1), FY 2000 and partial FY 2001 data indicate fairly stable types of MC and SDM involvement. Of every 10 clearances, SDMs are deciding about 2 cases on their own and use MCs as consultants and/or for forms assistance in about 3 cases. Four clearances require MC input and one clearance does not involve the SDM at all.

When we asked if MCs are involved in the most difficult cases (e.g., cases having multiple impairments), we were told the Office of Quality Assurance and Performance Assessment (OQA) is collecting this information and it would be reported sometime after the Prototype interim report is released. However, the interim report does have some data. SDMs appeared to use the MCs for more difficult cases because certain impairments have a higher MC involvement than others. For example, SDMs used MCs in 50 percent of back disorders and 74 percent of chronic pulmonary heart cases.

Claimant conference (usually via telephone) provides an opportunity for the claimant to talk with the SDM after the SDM has made a less than fully favorable decision. The SDM discusses the information in the claimant’s file and assures all pertinent evidence has been obtained. In addition, the SDM helps the claimant to understand SSA’s disability process, the requirements for entitlement, and answers questions.

The Office of Disability told us that a number of DDSs have voluntarily instituted first-day calls; that is, having the SDM contact the claimant by telephone immediately after receiving the case. The purpose of the first-day call is to ensure the claimant has submitted all pertinent medical evidence before the SDM adjudicates the case and to decrease the amount of time the SDM devotes to the claimant conference later. A first-day call is used in addition to the claimant conference.

Because the claimant conference consumes more examiner time compared to the current process, SSA is testing three alternative claimant conference processes in Alabama, Michigan, Missouri and Pennsylvania. Refinements have been made to claimant conference procedures to make it less formal and more efficient. In addition, first-day calls are optional in the Michigan and Missouri pilots, while Alabama and Pennsylvania have made first-day calls mandatory. The interim evaluation report on these alternative processes was to be released in November 2001.

Claimants who receive an unfavorable decision (a denial) are offered a claimant conference via telephone or face-to-face. In the Full Process Model (FPM), 56 percent of the claimants who would have been denied at the initial level took the opportunity to participate in the pre-decision interview, the predecessor of claimant conference. By May 2001, overall claimant participation increased to 64 percent in the Prototype.

Similar to the FPM test, a difference in rates of participation exists between title II (Disability Insurance) and title XVI (Supplemental Security Income) claimants. Title II claimants still have the highest participation rate (72 percent) as compared to title XVI (SSI) claimants (61 percent). Both rates are improvements over the FPM rates of 65 percent for title II and 49 percent for title XVI claimants.

In a recent OQA customer satisfaction survey of claimants whose cases are part of the longitudinal database, a majority of those who participated in the claimant conference rated their satisfaction with the SDM’s performance as excellent, very good, or good (E/VG/G). Below are some performance items and the E/VG/G ratings. Predictably, those who were awarded disability benefits ranked SDM performance from 9 percent to 27 percent higher than those who were denied benefits.

One of the goals of the Prototype process is to make more allowances earlier in the disability determination process. According to the Commissioner, allowances were made several months earlier in the disability determination process due to the elimination of the reconsideration level. The Prototype interim report stated that allowances are made 135 days earlier on the average.

In the Prototype interim report, evaluation data indicated a small difference between the Prototype and comparison States in regard to allowance rates - 40.4 percent total allowance rate for Prototype States, as compared to 39.8 percent for comparison States. Since FY 1999, the Prototype allowance rate has increased by 1.5 percent from 38.9 percent to 40.4 percent. Thus, in the Prototype more allowances are being made earlier in the process.

In regard to additional allowances made after the claimant conference, we used the number of allowances after the claimant conference offer and the number of clearances from the Prototype website monitoring data for FY 2000 to determine approximately 3.8 percent were additional allowances.

Average initial case processing times have increased for DDSs nationwide, but especially for Prototype DDSs as the new disability process is implemented in these States. In Prototype States initial processing time has increased by 23 percent (from 81.8 to 100.4 days) over the FY 1999 level, whereas in comparison States the increase is 10 percent (78 to 86 days). Prototype denial decisions (110 days) take 23 days23 days longer to process than allowance decisions (87 days) as compared to about 11 days11 days longer in the comparison States.

Prototype production-per-workyear (PPWY) data indicate that the number of case clearances processed per workyear has decreased by 13.4 percent. For Prototype States, PPWY fell from 253 in FY 1999 to 219 by March 2001. In contrast, for nonPrototype States, PPWY has been increasing from 262 in FY 1999 to 282 by March 2001.

Prototype pending cases now age 70 days on the average, the range being 47 to 102 days. As of July 27, 2001, pending cases in Prototype States number over 126,000, and the backlog of work is growing. The number of "weeks work pending" was 10.2 in FY 1999 and 15.7 in FY 2000. In comparison, for nonPrototype States it was 11.3 and 12 weeks, respectively.

The Office of Disability told us that lower performance by Prototype DDSs was expected. With the changes in disability operations required by the Prototype and Process Unification initiatives, slower processing times and lower productivity are the results of, for example, the additions of claimant conference and rationales to disability operations.

However, the goals of customer service and a more efficient, timely process do not seem to be well-served by the Prototype process. This scenario is particularly troublesome given that SSA’s Office of the Actuary has projected a 63.6 percent increase in the number of SSA disability beneficiaries by 2010. These data may indicate that more customers will wait longer for disability decisions at the DDS level in the future.

Anecdotally, Office of Disability officials have stated that processing time of disability cases has increased because more processing time is required by claimant conference and rationales of disability decisions. In testimony before the Committee on Ways and Means, Subcommittee on Social Security, Acting Commissioner Larry Massanari gave more specific information, stating that processing time for a denial decision takes 20 days longer, mostly due to claimant conferences. In the FY 2002 SSA budget submission, the Office of Disability made the following assumptions about the effect of the new disability determination process on time spent on each initial disability case:

• rationales will require an additional 75 minutes;

• claimant conference will require 35 minutes for each potential denial;

• claimant telephone calls and interruptions will add 10 minutes;

• the learning curve will add 45 minutes and gradually diminish over a year.

As a note of caution, the Office of Disability is not collecting data on how much time is actually added to case processing time by the claimant conference or rationales, so we do not have verification of the actual time per case these Prototype features require.

SSA has taken some action in regard to decreasing the time spent on rationales. According to the Office of Disability, a major consumer of the SDM’s time has been ensuring that all relevant Process Unification issues have been addressed in the rationale for each disability decision. Because SSA disagrees with the claimant in a denial decision, it is especially important to have a well-documented and logical rationale that fully explains the decision. It is less important in an allowance decision, since SSA agrees with the claimant and the claimant will not be filing an appeal. In December 2000, recognizing the need to lessen some of the unnecessary documentation and explanation, the Office of Disability issued instructions to DDS administrators to discontinue formal rationales for fully favorable decisions.

We asked the Office of Disability if the rationales prepared by the DDSs are helpful to OHA in terms of saving OHA some processing time. The Office of Disability is not collecting information in this area and could not answer our questions.

In a notice of proposed rule making published in the Federal Register, January 2001, SSA stated that no program savings would result from implementing the (Prototype) process nationwide. From FY 2001 to FY 2005, program outlays would increase for title II by $2.58 billion and Medicare costs would increase $277 million. From FY 2001 to FY 2004, program outlays would also increase for title XVI by $303 million, and additional Federal Medicaid costs would be $473 million while State costs would be $357 million. Finally, SSA stated there would be "some administrative costs associated with the transition to these rules" without specifying an amount or range of administrative costs.

When the Prototype was initially implemented in October 1999, SSA started out with the hypothesis that the resources saved from eliminating the reconsideration level in the Prototype DDSs would offset the costs of processing disability claims in the new disability determination process. This has not happened. In fact, the Prototype rollout set for April 2002 has been delayed indefinitely until more data on the impact of costs are available from the Prototype evaluation (and, more recently because SSA is concerned about the significant increase in the number of initial claims appealed to OHA).

In a letter to DDS administrators in May 2001, the Office of Disability stated one reason for increased program costs is some of the allowances made at the DDS level under

Prototype would not have been appealed and allowed at OHA under the old process.

According to the SSA officials we interviewed, other factors have contributed to administrative costs, such as:

• funding has been insufficient to pay for receipts;
• the continued costs of processing pipeline cases (i.e., cases under the old process with the reconsideration level);
• high attrition among disability examiners (DE);
• variance of different pay levels and benefits for DEs by State;
• Process Unification issues (e.g., the increased emphasis on enhanced explanation and documentation of treating sources); and
• the increased emphasis on claimant conference.

In the Prototype interim report, additional factors were mentioned as responsible for increased administrative costs:

• the learning curve for DEs was longer than anticipated;

• pre-Prototype PPWY in FY 1999 for Prototype States was below the national average, indicating production problems unrelated to Prototype; and

• resources are being funneled into the backlog because pendings are too high.

During our discussions, SSA could not give us any specific information on the administrative costs or the program costs of operating the Prototype. While awaiting the results of the analysis of costs (part of the Prototype evaluation), SSA officials created the Prototype Process Refinements Team to "refine" policies and procedures, change regulatory and/or statutory language, and make other adjustments to the disability process to make it more efficient.

For DEs and other DDS staff, the implementation of the Prototype process (along with Process Unification) represents a period of transition and change. One of the effects of change on the DDS staff can be seen in the DE attrition data for FY 1997 (pre-Prototype baseline) and FYs 2000 and 2001 (the first 2 years of Prototype operation), and the DDS staff attrition data for FYs 1999 and 2000. As a reflector of change, attrition rates have increased.

The data in Chart 2 above indicate 6 of the 10 Prototype States had higher DE attrition rates in FYs 2000 and 2001 than they did in FY 1997 (Alabama, California, Colorado, New Hampshire, New York, and Pennsylvania). Nationally, the DE attrition rate increased nearly 3 percent from 10.2 percent in FY 1997 to 13.1 percent in FYs 2000 and 2001. Also, 6 Prototype States (Alabama, Alaska, Colorado, Louisiana, Missouri, and New Hampshire) were above the national average DE attrition rate in FY 2000, and 5 of the same 6 Prototype States (except for Alaska) were above the national average in FY 2001 as well.

In Chart 3 below, data on the 7 comparison States show 3 States (Georgia, Kansas, and Utah) with an increase in attrition in FYs 2000 and 2001 as compared to FY 1997,FY 1997, and 1 state (Vermont) with an increase in FY 2001 only. Two comparison States (Georgia and Kansas) were above the national average attrition rate in FY 2000, and 3 StatesFY 2000, and 3 States (Georgia, Utah, and Vermont) were above the average of 13.1 percent in FY 2001.FY 2001.

For most DDSs nationwide, retention of qualified DEs seems to be problematic. It takes about 2 years for a DE to become proficient in the job. At present, more than 50 percent of the current DEs have less than 2 years of experience in the job. Anecdotal reasons for why examiners leave are retirement, dealing with the public via claimant conference, better paying jobs elsewhere and difficulty of the work at the DDS. DEs in one State may leave to work for a contiguous State offering higher salary and benefits to examiners. Or, DEs may leave the DDS to take another State job at equal or higher salary with less responsibility.

Chart 4 shows DDS staff attrition rates for FYs 1999 and 2000. Eight of the 10 Prototype States had higher DDS staff attrition rates in FY 2000 than they did in FY 1999.FY 1999. Overall, the national DDS staff attrition rate increased from 10.5 to 11.6 percent11.6 percent from one year to the next. In FY 2000, 6six Prototype States exceeded the national DDS staff attrition rate.

SSA reports in the Prototype interim report that site visits to Prototype DDSs revealed DDS staff were concerned about several matters such as:

• the long learning curve to create detailed rationales, and dealing with issues of credibility, symptoms and pain, and medical source opinion in rationales;
• implementing many changes at the same time;
• the additional processing time added by claimant conference; and
• increasing caseloads.

Whether these matters contributed directly to the attrition rate we cannot be certain, since no current data exist to confirm this hypothesis. However, given that the Prototype and Process Unification initiatives constitute a very different way of doing business, these concerns at Prototype DDSs may generate a climate conducive to attrition.

With the advent of the Prototype and Process Unification initiatives, the expectations for examiners in particular have increased significantly. Examiners are expected to go through more training, to develop skills to deal with the public, to write analytically for rationales, and to apply a wide range of knowledge about the medical and vocational aspects of disability and complex Process Unification issues. Under Prototype operations, the job of examiner has become more rigorous and more highly skilled. In this respect, not all current examiners can become SDMs. See pages 6-8 in this report for further discussion.

According to information from the Office of Disability, overall accuracy rates have not increased, but they have remained stable. In FY 1999 the overall accuracy rate for all clearances was 94 percent.94 percent. In FY 2001 (through March) the rate was 94.1 percent.

In the Prototype interim draft, the reported return rates indicated allowance accuracy has improved slightly. In FY 1999 the return rate for allowances was 3.8 percent; the current return rate is 3.4 percent. Denial accuracy is still a problem for Prototype DDSs and for most DDSs nationwide. In FY 1999 the denial return rate was 7.4 percent; the current rate is 7.6 percent.

Quality Assurance (QA)

Initially, improvement of the Social Security Administration’s (SSA) disability QA process was conceptualized as (1) developing in-line quality assurance processes, and (2) developing a single, end-of-line review mechanism that looks at the whole adjudicative process. In-line quality assurance processes include the development and implementation of procedures and tools such as access to medical/vocational experts, regular training to learn new skills and maintain learned skills, mentoring by experienced employees, peer discussions and review of work products, and immediate feedback on case development and adjudication. The end-of-line quality assurance mechanism, separate from pre-effectuation reviews, would include a final review of initial and hearing-level decisions and a feedback process having the same standard of review for all cases, e.g., preponderance of evidence vs. substantial evidence.

The Disability Program Redesign Team originally had the lead on this initiative and formed two task forces: an "in-line" QA task force to look at controls and feedback mechanisms and an "end-of-line" QA task force to develop an end-line measurement for quality. The "in-line" and "end-of-line" task forces could not reach any clear consensus on changes and measurements for the QA process, and no presentation was made to the Commissioner representing the work of these task forces.

In 1998 the lead component for the QA initiative switched to the Deputy Commissioner for Finance, Assessment and Management, and the central question became, "How do you define and measure accuracy?"

The second report was on benchmarks. The October 19, 2000, report described the quality assurance process in two public agencies (Veterans Administration and the Food Stamp Administration) and two private businesses (UNUM/Provident Insurance Company and the Colorado Foundation for Medical Care). These benchmarks were not a one-to-one correspondence with SSA’s experience, but they did offer SSA a guide to best practices of other public and private organizations and some ideas as to the applicability of the practices.

The draft final report, originally issued on August 30, 2000, combined the findings of the first two reports along with Lewin and PEM Associates’ recommendations for an advanced quality management system. After an SSA internal review, the contractor was asked to more clearly structure the report to delineate long-term and short-term options for an effective QA system as intended in the statement of work for this contract. The contractor released the revised version on November 24, 2000, and SSA circulated the report for review and comment.

The final report was issued March 16, 2001. Lewin and PEM Associates told SSA that tinkering or retooling or adding resources would not move SSA toward its quality improvement goals. Instead, SSA needs "…to adopt a broad, modern view of quality management that involves efforts outside of OQA and the current quality assurance process." The report includes short-term and long-term options in eight areas: Leadership and Organization, Performance Management System, Promoting a Quality Culture, Quality Control, Performance Monitoring Systems, Federal-State Relationships, Initial Disability Determination Process, and Appellate Process.

Disability Claims Manager (DCM)

• Overall the DCM processing time was faster. DCM median processing time for initial claim allowances was about 10 days faster for title II and 6 days slower for title XVI as compared to the control group. For denials, median processing time was 14 days faster for title II and 13 days faster for title XVI initial claims than the control group processing times.

• Initial allowance rate and the cumulative rate through the reconsideration level were about the same for the DCM group (initial 43.8 percent and reconsideration 46.9 percent) and the control group (initial 43.6 percent and reconsideration 46.6 percent).

• 9.1 percent more claimants in the DCM group requested a hearing as compared to the control group.

• DCM allowance (96.9 percent) and denial accuracy (90.1 percent) rates were comparable to the control group accuracy rates (96.8 percent allowance accuracy and 93.3 percent denial accuracy).

• The DCM denial accuracy rate (90.1 percent) was below the regulatory threshold of 90.6 percent.

• DCM productivity ranged from 14 percent less to 8 percent more as compared to the current process. The variation in these figures is due to two different productivity measurements - for the Disability Determination Service (DDS), Production Per Work Year (PPWY); for the field office, Work Units Per Work Year. Lewin helped SSA to adjust the productivity formulas for field offices and DDSs to allow a comparison of the DCM to the current process.

• Customer satisfaction (excellent, very good, good responses) of denied DCM claimants was significantly higher than the group surveyed by the Fiscal Year (FY) 2000 Market Measurement Program Survey of initial disability applicants (not the control group), 68 percent versus 55 percent. For allowed DCM claimants, customer satisfaction was about the same (94 percent) compared to the Market Measurement Program survey respondents (91 percent).

• DCMs rated their job satisfaction in the DCM position higher than their former positions. In Phase I, 82.4 percent of DCMs said their job satisfaction improved, as was the case in Phase II (83.1 percent).

• The overall DCM attrition rate was significantly higher (17 percent) than the combined DE and claims representative attrition rate (10.1 percent) for the 15 States in the test. During the 3 months before the DCM test ended (September to November 2000), SSA reported two-thirds of the attrition occurred. SSA does not know if attrition was affected by the end of the formal evaluation period in November, even though DCM operations continued through June 2001.

• No significant differences between DCMs in the Prototype process and the control group were noted on such indicators as pre-effectuation review accuracy, median processing time and initial allowance rate, as shown in table 3 below:

• Implementation of the tested DCM must be legislated. The DCM in any other form requires agreements between Federal and State levels.

Process Unification

• SSR 96-1p: "Acquiescence policy." Policy on how SSA acquiesces when a final circuit court law conflicts with SSA policy.

• SSR 96-2p: "Giving Controlling Weight to Treating Source Medical Opinions." Policy on giving deference to a treating physician’s medical opinion on the nature and severity of an impairment when the opinion is not inconsistent with other substantial evidence in the claimant’s file and the opinion is supported by medically acceptable diagnostic techniques.

• SSR 96-3p: "Considering Allegations of Pain and Other Symptoms in Determining Whether a Medically Determinable Impairment is Severe." Policy on determining the severity of an impairment at step 2 of the sequential evaluation process.

• SSR 96-4p: "Symptoms, Medically Determinable Physical and Mental Impairments, and Exertional and Nonexertional Limitations." Policy on determining a mental or physical impairment by medical signs and laboratory results and the type of limitations of function restricting work ability.

• SSR 96-5p: "Medical Source Opinions on Issues Reserved to the Commissioner." Policy on issues such as whether an individual's impairment(s) meets or is equivalent in severity to the requirements of any impairment(s) in the listings, what an individual's residual functional capacity (RFC) is, whether an individual's RFC prevents him or her from doing past relevant work, and how the vocational factors of age, education, and work experience apply.

• SSR 96-6p: "Consideration of Administrative Findings of Fact by State Agency Medical and Psychological Consultants and Other Program Physicians and Psychologists at the Administrative Law Judge (ALJ) and Appeals Council (AC) Levels of Administrative Review; Medical Equivalence." Policy on using DDS-level medical and psychological findings at the ALJ and AC levels.

• SSR 96-7p: "Evaluation of Symptoms in Disability Claims: Assessing the Credibility of an Individual’s Statements about Pain or Other Symptoms." Policy on substantiating a claimant’s statements (or determining credibility) about pain and other symptoms with medical signs and laboratory findings or, in the absence of these, with the entire case record.

• SSR 96-8p: "Assessing Residual Functional Capacity (RFC) in Initial Claims." Policy on determining RFC in steps 4 and 5 in the sequential evaluation process.

• SSR 96-9p: "Determining Capability to Do Other Work—Implications of a RFC for Less Than a Full Range of Sedentary Work." Policy on the impact of a RFC assessment for less than a full range of sedentary work on an individual's ability to do other work.

Traditionally, the hearings offices have relied on laws and regulations, and the Disability Determination Services (DDS) used the Program Operations Manual System (POMS) instructions as the basis for adjudicating disability claims. SSA told us that a change in the wording of policy as it undergoes language changes from regulations to instructions in POMS causes different interpretations of policy at the DDS level and the hearings office level.

SSA began the implementation of a single presentation of policy binding on all levels of decision makers in 1995 when SSA started issuing new adjudicative policy guidelines in the same wording.

Issued by the Commissioner, SSRs are court decisions, policy statements or legal opinions. SSRs do not have the effect of law or regulations, but are regarded as binding policy on all SSA components. The 9 SSRs issued in July 1996, known as the Process Unification rulings, clarified policies contributing to inconsistent decision making. SSA has started the practice of printing the SSRs in POMS.

Most of these SSRs have significantly impacted DDS examiners and the adjudication process. An SSA official told us that, when the SSRs were issued, SSA knew they would have a great impact on resources at the DDS and hearings office levels; e.g., that productivity would decrease and that changes would have to be phased into disability operations. The Office of Disability told us that using the Process Unification rulings (e.g., assessing credibility, assessing RFC, and resolving conflicts in medical evidence) in writing rationales has increased claim processing time and decreased productivity. One SSA official told us, "Process Unification needs more resources, or the rules need to change."

The report on the Philadelphia workgroup’s recommendations addresses several issues raised by the application of Process Unification rulings to case adjudication. Some issues have simple solutions, such as "provide examples of Process Unification documentation." The resolution of other issues will require revisions in SSRs, or changes in regulations and legislative language. Implementation work on these recommendations and other recommendations is spearheaded by the Process Redesign Refinement Team - comprised of employees from the DDSs, Regional offices, Disability Quality Branches (DQB), Office of Hearings and Appeals (OHA), and SSA Headquarters - and the 30-Day Workgroup.

In terms of resolving differences in decision making between the DDSs and ALJs, SSA told us that these differences may not entirely disappear due to the nature of the work, which involves judgment and discretion on the part of DDS decision makers and ALJs. Some of the differences involve questions surrounding the use of subjective evidence; e.g., a claimant’s description of pain. Some of the differences involve varying standards of evidence. For example, the DQB uses a "preponderance of evidence" standard, while OHA adjudicators use the "substantial evidence" standard.

• Complete documentation and detailed explanation (rationale)

• Remand selected hearing cases

• Quality review of hearing decisions

The final feature of the Process Unification initiative is SSA’s commitment to provide training to adjudicators at all levels of the disability process in several areas - assessment of symptoms, treatment of opinion evidence, credibility, and assessment of RFC. In 1997 SSA began the training regimen with nationwide training for 15,000 disability adjudicators, including DDS examiners, physicians, ALJs, and quality reviewers.

In the January 2001 disability report issued by SSA, the Commissioner indicated there were several key training activities subsequent to the training for 15,000 disability adjudicators, such as:

• Over 60 disability training programs were presented via interactive video training in the last 3 years to disability adjudicators across the country on Process Unification issues and on new or revised disability benefit processes.
  
  
• The national Disability Examiners Basic Training Program was updated in Fiscal Year 2000, incorporating Process Unification principles and the latest operating procedures.
  
  
• In March 2000, a Disability Adjudicators Trainers Workshop was attended by more than 100 DDS trainers to share training material, techniques and best practices.

• Two new training groups were established. Disability Training Steering Committee consists of representatives from various SSA and DDS components who are to assess the role and functions of disability adjudicators and recommend a training plan to achieve SSA goals. The National Disability Training Cadre will create and deliver nationwide training to all disability program personnel.

Hearings Process Improvement (HPI)

The HPI Plan was created to address problems with the structure and work process of hearings offices (HO). The pre-HPI work process in HOs involved numerous handoffs and a high degree of work specialization. In the HO structure there was no first-line manager to directly oversee the entire workflow from case receipt to disposition. Each staff group focused on a specific aspect of production rather than on processing the case timely and with the customer in mind. Finally, inadequate automation and management information was a major impediment to an efficient and effective system of case processing. The following is a summary of the HPI Plan key elements, which address these problems.

National Workflow Model. A key strategy to reduce processing time, improve productivity, and effect fewer case handoffs, is the implementation of a national workflow model. The model will be standardized throughout all offices with some variation permitted. Up-front screening, development, and routing will ensure that certain cases (e.g., possible dismissals and on-the-record cases) are assigned immediately to a legal analyst for review, and that cases are selected and assigned to the processing group based on established criteria. Standardized automation procedures such as development and maintenance of automated notices, automated tickler follow up, case tracking and enhanced decisional support are extremely important for the success of the national model workflow.

Processing Time Benchmarks. Benchmarks will be established for the overall hearing process and tasks within the process. Benchmarks will limit elapsed time before follow-up on the claimant's response. If the claimant responds within the time limit, a development analysis will ensue immediately. If the claimant fails to respond, the claimant will be contacted quickly.

Processing time benchmarks for specific development actions (e.g., receipt of information) will be included in a proposed electronic Development Calendar. As a date approaches, the Development Calendar Tickler Reports will pinpoint cases to follow up when required information has not been received.

Pre-Hearing Conference. Full development of cases is not always accomplished prior to a hearing. Ensuring that each scheduled case is ready for hearing is necessary to improve the process. The new process provides opportunities for claimants and their representatives to participate in the early completion of case development. Each case’s development needs will be identified early by an analyst, and the claimant will be offered the opportunity for a developmental conference or a pre-hearing conference (PHC), as appropriate. The developmental conference allows the claimant to submit more current or additional evidence and conveys information to the claimant about the hearings process, development of the claimant’s case, the right to and availability of representation, and how the claimant can help to speed processing of the claim. A formal PHC may be offered to the claimant to pinpoint issues to be adjudicated, or the PHC may result in an on-the-record decision, which eliminates the need for a hearing. When case development is complete, cases will be certified as "ready to hear."

Under the new hearings office structure, a Hearing Office Director (HOD) position will be created, under the general direction of the Chief Administrative Law Judge, to manage the HO workflow and to ensure the timely and effective processing of cases. The HO will be organized into "corps units," each of these processing groups serving about four judges. Each work group will be organized under a Group Supervisor responsible for the management of the group's workload. The Group Supervisor will work with the Legal Advisor who will provide legal/technical assistance to the group, as well as to the HOD and the judges. The Legal Advisor will also review possible on-the-record decisions and dismissals, and assist in writing more complex decisions.

The success of HPI depends heavily on enhanced automation, electronic data collection and analysis. To function effectively, the new process must have the support of automation for monitoring workflow, tracking and tickling of case processing and development steps, transferring case information, scheduling, and timely management information reports. This degree of management information support is not presently in place.

[SSA also provided general and technical comments that we incorporated into the report as appropriate.]

Bill Fernandez, Director, Western Audit Division, (510) 970-1739
Mark Bailey, Deputy Director, (816) 936-5591

In addition to those named above:

Carol Cockrell, Program Analyst
Sherry Colwell, Lead Auditor
Cheryl Robinson, Writer-Editor