The Performance of Social Security Administration Appeals Offices
Good morning, Mr. Chairman, Mr. Johnson, and members of the Subcommittee. Before I begin, I want to express the condolences of my entire organization at the loss of Ms. Tubbs-Jones. As you know, the Congresswoman had been a long-time member of this Subcommittee, and a long-time supporter and friend of the Office of the Inspector General. I met with her earlier this summer and we enjoyed a frank exchange of ideas directed toward a common goal-improving Social Security programs for her constituency and for all Americans. I've appeared before this Subcommittee many times during my tenure as Inspector General, and on almost every occasion, she held my feet to the fire, driven by her own commitment to public service and to the people who elected her. Like you, and like the people of Ohio, I lost a friend. The thoughts and prayers of all 600 employees of the Office of the Inspector General are with her loved ones.
The best way to serve her memory is by doing exactly what we’re doing here today-standing together and looking under the hood of the Social Security disability appeals process to find ways to make the engine run more smoothly. The disability appeals backlog is unacceptable to the Social Security Administration (SSA), to you, and to me, just as it is unacceptable to the American public and to everyone at this witness table. This hearing by no means represents our first attempt to make the necessary repairs, nor will it be our last. I believe, however, that it represents an important step forward, as the work requested by this Subcommittee and recently completed by my office sheds some new light on the challenges confronting us.
By way of background, it is important to understand that the backlog is not the result of a lack of dedication or commitment on the part of SSA or any of its employees, nor of the Administrative Law Judge (ALJ) corps, though it falls on all of these parties to join in seeking solutions. In 2004, looking at hearing office factors that contributed to the increasing backlog, my office found that although the number of dispositions had increased-ALJs were processing more cases than ever before-the number of incoming appeals was growing even faster, leading almost inevitably to longer processing times and an increased backlog. And the phenomenon has continued. In 2001, the Office of Disability Adjudication and Review’s (ODAR) average processing time for an appeal was 308 days. Now, despite all of our efforts, the average processing time is 505 days.
Earlier this year, we conducted an audit that focused specifically on ALJ productivity. While we recognized then-as we do today-that not all of the responsibility for the backlog can be laid at the feet of the ALJs, understanding the challenge and seeking a solution is not possible without understanding the role of the ALJ in the process.
We found significant discrepancies in ALJ productivity in that audit. We also found, however, that processing delays and increases in the appeals backlog were partially attributable to hearing office staffing, use of management information, and other issues not directly related to the ALJs themselves.
While we were reporting these hearing office-based findings, the Subcommittee requested that we undertake a separate study on the key role that the hearing office plays in the efficiency of the disability appeals process. Specifically, you asked that we consider ALJ case disposition statistics, but also that we examine case processing times, the reasons for variances among hearing offices and ALJs, ODAR’s management of ALJ performance, and SSA’s management initiatives aimed at reducing the backlog and improving processing time.
To accomplish this, we conducted the most thorough review we have ever undertaken in this area. We visited 49 of ODAR’s 141 hearing offices across the country. In each of these 49 offices, we interviewed the Hearing Office Chief ALJ (HOCALJ), one high- or low-producing ALJ, a mid-producing ALJ, the Hearing Office Director, a senior attorney-advisor, and a senior case technician. We interviewed Judge Cristaudo, SSA’s Chief ALJ, 9 Regional Chief ALJs, a total of 143 ALJs, and a total of 146 hearing office staff.
We looked at management tools and practices, including disciplinary actions taken against ALJs for performance issues, and we studied 37 initiatives that SSA has undertaken to reduce the backlog. Our findings, organized by the four specific subject areas set out in your request, follow.
ALJ Case Disposition Statistics
To analyze ALJ case disposition data, we looked at the case disposition statistics of all ALJs who issued dispositions during the years in question. This included full-time and part-time ALJs, new ALJs, and ALJs who may have retired, separated, resigned, or died during the year. This also included ALJs with union or management duties, of whom fewer cases might be expected, as long as they issued at least one disposition. In other words, rather than use the traditional government notion of “full-time equivalents,” we looked at every ALJ who issued even one case disposition in the given year.
On the other side of the equation, we considered all case dispositions, including cases remanded to the appropriate disability determination services (DDS) office for further processing or consideration of an allowance.
Using these definitions, we found that the average number of case dispositions per year per ALJ had increased by 13 percent between fiscal year 2005 and fiscal year 2007. Specifically, the ALJ corps averaged 421 case dispositions in 2005, and improved to an average of 474 case dispositions in 2007. This 2007 average saw 1,155 ALJs issue a total of 547,951 dispositions. Discounting approximately the highest and lowest five percent of ALJs to eliminate statistical anomalies, the number of dispositions issued by the ALJs in our study (which included both fully- and partially-available adjudicators) ranged from 101 to 800 cases. Later in my statement, I address some of the reasons for this wide range.
Case Processing Time
The numbers that matter most to the public have nothing to do with ALJ productivity or hearing office practices. Rather, the understandable concern of any disability appellant is how long it will take SSA to render a decision on his or her appeal.
Looking at the same period, the average case processing time in fiscal year 2005 was 443 days. By fiscal year 2007, the average processing time had increased 16 percent, to 512 days. This, despite a 13 percent improvement in ALJ case disposition numbers. Again discounting for statistical anomalies, for ninety percent of the ALJs in our study, the average processing time per ALJ ranged from an average of 301 to 700 days.
Understandably, SSA attributes much of the increase in case processing time to increases in the numbers of appeals filed and limitations on resources necessary to process these appeals. This is true, but as you were aware when you requested this review, investment in improving this process must be made wisely and carefully. While hiring additional ALJs is a necessary component of improvement, that alone will not resolve the matter.
It is also worth noting that of ODAR’s 141 hearing offices, 22 of them, or 16 percent, had average case processing times that exceeded the national average (512 days) by 100 or more days. This suggests that both ALJ productivity and hearing office practices play a role in processing delays.
Reasons for Variances Among Hearing Offices and ALJs
We identified eight major factors that contribute to the wide variances described above among hearing offices and among ALJs.
Valid and/or Immutable Factors
As stated earlier, a thorough study required that we look at all ALJs that issued dispositions in a given year. This meant that we could not take into account-in our overall analysis-ALJs with good reason for issuing relatively few dispositions. Looking beyond the initial review, however, reveals that in many cases, what appear to be lower-producing ALJs are not cause for concern.
We looked again at the 95 ALJs in our study of fiscal year 2007 who issued fewer than 200 case dispositions. We found that of these 95, one was Judge Nancy Griswold, the Deputy Chief ALJ, who certainly had other issues occupying her time. Similarly, five of these 95 ALJs were Regional Chief ALJs.
Another 13 of these 95 ALJs were new to their jobs (and thus had a significant learning curve), were part-time employees, or were on extended leave during the year. And another 54 of these ALJs either retired, separated, resigned, or passed away during fiscal year 2007. This left 22 ALJs who produced fewer than 200 dispositions. Ten of these 22 ALJs were union officials who, under the collective bargaining agreement, had officially authorized union responsibilities. We interviewed the ten union officials as part of our study.
We then interviewed the twelve remaining ALJs, each of whom issued between 150 and 200 dispositions during fiscal year 2007. The reasons they cited for their disposition numbers are incorporated in our report, and in this testimony.
Through our interviews, we found that internal factors-unquantifiable factors internal to each ALJ-were significant contributors with respect to disposition productivity. In fact, our interviews with Regional Chief ALJs (RCALJ) revealed that work ethic and motivation were one of the main factors that contributed to high or low productivity. One of these interviews even revealed an ALJ who remained unmotivated despite oral and written counseling, a written directive, and a reprimand.
One RCALJ told us “Some ALJs process fewer cases than expected due to a lack of motivation.” Another stated that “Some ALJs are not motivated to process more cases or are stuck in a time when fewer dispositions were expected.”
Since, however, work ethic and motivation-as well as other internal factors-are particular to each ALJ and cannot be quantified, our ability to study this factor’s precise effect on processing time and on the backlog is limited.
We consistently heard from ALJs and hearing office staff that DDS disparities were a significant factor with respect to hearing office performance and processing times. As one RCALJ stated, “Poor quality cases from the DDS level can cause some ALJs to process fewer cases.”
Hearing offices likely receive more hearing requests if they are located in states with DDSs that deny more initial claims. (ALJs in these hearing offices may have higher favorable rates.) By the same token, the length of time that ALJs spend reviewing cases prior to a hearing may be affected by the extent to which the DDS developed the case-DDSs that do more work may make it easier for ALJs to process cases more quickly.
The support staff in SSA hearing offices conduct initial case screening and preparation, maintain the case control system, conduct pre-hearing analysis, develop evidence, schedule ALJ hearings, prepare notices and decisions, and perform various other functions in support of the appeals process. As we recognized both in our February audit and in this one, insufficient staffing appears to be a factor in ALJ and hearing office performance and case processing times.
We found that hearing offices with a staff ratio higher than the national average of 4.46 staff per ALJ were likely to have higher-producing ALJs. Specifically, we found that more than half (52 percent) of the higher-producing ALJs we interviewed were in offices with staff ratios higher than that average, but only 17 percent of the lower-producing ALJs we interviewed were in hearing offices with above-average staffing levels. Similarly, 63 percent of the hearing offices ranked by ODAR as being in the top half nationwide for productivity had a staff ratio higher than 4.46 staff per ALJ, while only 38 percent of the hearing offices ranked in the bottom half were staffed above that level.
Our conclusion that staffing was a key factor in hearing office productivity was confirmed by the fact that all 48 hearing office directors we interviewed stated that staff ratio had a significant impact, and hearing office staff in 39 of the 49 offices we visited told us that more staff was needed.
“Support staff ratios have a significant impact on productivity and processing times,” said one RCALJ. “Hearing offices often over-burden the strongest employees which often leads to the best staff leaving the office and a demoralization of the office. Further, it is difficult to meet timeliness goals with limited staff. If one staff person is gone, there is often no backup.”
Typically, ALJs provide hearing office staff with the number of hearings the ALJ would like to have scheduled three months in advance of the period being scheduled. We found, however, that 55 percent of the lower-producing ALJs sometimes did not have as many hearings scheduled as they requested, generally due to staffing levels that were insufficient to support preparation for that many hearings. Of the ALJs who told us this, 39 percent went on to state that this was a regular occurrence. Moreover, we learned that in offices where this was a problem, most had staff ratios below the national average of 4.46 staff per ALJ.
In contrast, only 23 percent of the higher-producing ALJs told us that they regularly had fewer hearings scheduled than they requested, though they, too, pointed at insufficient staffing levels as the cause. Our interviews with hearing office directors and senior case technicians further confirmed this finding.
Said one low-producing ALJ, “While hiring more ALJs will help with hearing cases, the hearing office needs more trained staff.”
According to our study, higher-producing ALJs issued favorable decisions (decisions in which the appellant’s initial denial was reversed and the claimant was awarded benefits) in 72 percent of their dispositions, while lower-producing ALJs had a favorable rate of only 55 percent. Put another way, 65 percent of the higher-producing ALJs we studied had a favorable rate above the national average, while only 31 percent of the lower-producing ALJs had a favorable rate above the average.
This discrepancy is attributable to on-the-record decisions-cases reviewed by an ALJ in which the appellant is found to be eligible for benefits without need for a hearing. Higher-producing ALJs were more proactive in screening cases for on-the-record decisions, with 65 percent of them stating that they regularly screened cases for possible disposition in this fashion. Only 34 percent of the lower-producing ALJs stated that they regularly screened cases for on-the-record dispositions.
As one HOCALJ pointed out, “If goals are too high the corners get cut, and the easiest thing is to grant a case.”
Individual ALJ Preferences
We found that certain preferences of individual ALJs with respect to how cases were processed were indicators of higher or lower performance.
Case Preparation and Docketing
The amount of time spent reviewing a case prior to a hearing was a contributor to productivity: higher-producing ALJs spent an hour or less preparing a case, while lower-producing ALJs typically spent from three to eight hours.
Case docketing practices was also a factor. Higher-producing ALJs requested 10-50 hearings per week, while lower-producing ALJs requested between two and 30 hearings. Higher-producing ALJs were also more likely than lower-producing ALJs to schedule hearings before office staff prepares the file.
Length of Hearings
The length of hearings proved to be another indicator. Higher-producing ALJs stated that their hearings generally lasted less than an hour, while lower-producing ALJs stated that their hearings lasted from 30 to 90 minutes. Higher-producing ALJs also reached a decision more quickly, having reviewed the file beforehand and taken careful notes during the hearing.
Bench decisions-cases in which the ALJ rules in favor of the claimant during the hearing-are an indicator of higher ALJ performance. Only 14 percent of the lower-producing ALJs we interviewed issued bench decisions during fiscal year 2007, while 58 percent of the higher-producing ALJs utilized this practice.
By scheduling multiple cases involving unrepresented claimants for the same day and time, some hearing offices and ALJs are able to reduce their backlogs. Since cases involving unrepresented claimants are often dismissed (because the claimant does not appear) or postponed (because the claimant appears, only to decide that he or she wants representation), the rocket docket allows many hearing requests to be moved forward at the same time.
Time Spent Editing Decisions
The decision-editing process also slowed lower-producing ALJs, with 41 percent of them stating that they had substantial edits to more than half of the decisions prepared by their staff. None of the higher-producing ALJs we interviewed stated that they had such frequent edits.
In some areas, the ALJs and hearing offices do not always have control over factors that can cause delays. For example, under certain circumstances, an ALJ is required to obtain the testimony of medical or vocational experts, but in most cases, obtaining such expert testimony is discretionary. We found that 21 percent of the lower-producing ALJs used medical experts in more than half of their hearings, while only six percent of the higher producing ALJs fell into this category. Similarly, 72 percent of the lower-producing ALJs used vocational experts more than half the time, while 32 percent of the higher-producing ALJs did.
Like the use of experts, postponement can be mandatory, but is more often discretionary. In our study, 52 percent of the lower-producing ALJs had more than one-fourth of their hearings postponed. Only 32 percent of the higher-producing ALJs did.
Management of ALJ Productivity
We looked at the use of ODAR’s Case Processing and Management System (CPMS), and found that Hearing Office Chief ALJs (HOCALJ) use the system to monitor ALJ performance in varying degrees. Most of the HOCALJs we interviewed monitored the number of hearings that each ALJ in the office scheduled and met with ALJs who were scheduling low number of hearings. Five of the HOCALJs we interviewed actually approved each ALJ’s schedule, and it is worth noting that four of those five offices ranked in ODAR’s top 30 nationwide.
We also found, however, that fewer than half of HOCALJs were using CPMS to monitor bench decisions or on-the-record dispositions, methods described above as indicators of high ALJ productivity. The HOCALJs who did not use CPMS to monitor these types of cases stated that doing so would intrude upon an ALJ’s decision-making process.
Our study also looked at disciplinary actions taken against ALJs for performance issues. Only a few of the HOCALJs we interviewed stated that they would make recommendations for disciplinary action against ALJs for performance issues like low productivity. Among the reasons they cited was that such actions are difficult and time consuming.
Almost all of the 31 disciplinary actions initiated against 30 ALJs from fiscal year 2005 through June of 2008 were for conduct, not performance (there are two performance actions before the Merit Systems Protection Board), but Regional Chief ALJs we interviewed stated that they were beginning to address performance issues more than they had in the past. Counseling on performance issues also occurs, but is not tracked; the Regional Chief ALJs we interviewed indicated, however, that these also pertain mostly to conduct issues, not performance.
To address productivity, backlog, and processing time issues, the Commissioner implemented a four-pronged plan, to be achieved through 37 initiatives, many directly related to the factors I’ve discussed today. We believe that those that may have the most impact are the following:
This is the most direct initiative, and one which comports with our February 2008 audit on ALJ productivity. The Chief ALJ has requested that ALJs issue between 500 and 700 dispositions per year. In our February report, we stated that if ALJs were hired, and all ALJs completed 500 dispositions annually, the excess backlog would be eliminated by 2012. As of April, 49 percent of ALJs nationwide are on track to meet the Chief ALJ’s goal.
Hiring ALJs and Staff
In addition to 20 ALJs hired in fiscal year 2007, SSA has hired another 189 ALJs in fiscal year 2008. Further, ODAR is filling 230 staff positions in phases to balance staffing needs in each region. During Phase One, 92 immediate hires are being allocated to ODAR regional offices; during Phase Two, 138 hires are being allocated for distribution to the regions to backfill vacancies and balance staff ratios.
New Automation: Electronic Folder
In fiscal year 2007, ODAR transitioned from paper to electronic case folders (about 73 percent of folders were electronic as of March 2008). It was anticipated that this transition would bring with it a learning curve and period of adjustment, but some ALJs continue to assert that it is easier to use paper folders, and that the electronic folder slows the process. It is too early to assess the impact of the electronic folder on case processing times.
New Automation: ePulling
ePulling refers to customized software that is designed to facilitate the process of preparing cases for hearing. ePulling is underway on a pilot basis, with national rollout scheduled for fiscal year 2009. ODAR has estimated that it takes 3.5 hours to manually prepare an electronic folder for hearing, but that with ePulling, it will take only two hours (though additional staff time will still be needed after the ePulling process). My office has begun an audit on the ePulling pilot, and we will be happy to provide you with our findings when that work is complete.
DDS Informal Remand Project
Using profiles designed by SSA’s Office of Quality Performance, certain paper cases are sent back to DDS offices for a determination of whether a favorable decision can be issued without a hearing. The DDS staff, using overtime, reviews the case and if a fully favorable ruling can be issued, returns the case to SSA for processing and payment. If a favorable ruling cannot be issued, the DDS prepares the case for a priority hearing and returns it to the hearing office.
The results of this initiative are still being reviewed, and use of the same process for electronic files is also getting underway.
As part of the Commissioner’s plan, SSA intends to develop and implement a quality assurance program for the hearing process. The program will be rolled out in three phases. First will come a review of attorney adjudicator decisions; second, a review of decision drafts; and third, a review of cases with a hearing scheduled but not yet held. Reports will be issued and recommendations made based on SSA’s findings.
To date, SSA has reviewed 111 senior attorney adjudicator decisions and found them to be 95 percent accurate.
This statement summarizes the information presented in our Congressional Response Report, Administrative Law Judge and Hearing Office Performance. I believe the report, which is available on our website, provides Congress and SSA with a wide range of findings that may prove useful as we continue to work to ensure that disability applicants receive timely and accurate decisions on their claims. We have other work, both planned and underway, that focuses on various aspects of this challenge, and will result in recommendations to SSA for improvement.
Clearly, the hiring of additional ALJs and hearing office staff is the single most important step forward that can be-and now has been-taken. The prudent use of those resources, however, requires studies such as this, initiatives such as those put forward by the Commissioner, the support of the Chief ALJ, and the oversight of this Subcommittee. I thank you all for your interest, your concern, and your dedication. I would be happy to answer any questions.
Date: September 16, 2008
OIG Official: Patrick P. O'Carroll, Jr., Inspector General
Committee/Subcommittee: U.S. House Committee on Ways and Means, Subcommittee on Social Security