This is the accessible text file for GAO report number GAO-11-217R 
entitled 'Rail Transit: Reliability of FTA's Rail Accident Database' 
which was released on January 31, 2011. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

GAO-11-217R: 

United States Government Accountability Office: 
Washington, DC 20548: 

January 31, 2011: 

The Honorable Richard C. Shelby:
United States Senate: 

Subject: Rail Transit: Reliability of FTA's Rail Accident Database: 

Dear Senator Shelby, 

As you requested, we conducted a review of challenges associated with 
enhancing safety on major rail transit systems.[Footnote 1]During the 
course of that review, we assessed the quality of data that the 
Federal Transit Administration (FTA) collects and maintains in its 
State Safety Oversight (SSO) Rail Accident Database. FTA, an agency 
within the Department of Transportation (DOT), collects these safety 
data, including data on types of accidents and causes, from SSO 
agencies and the rail transit agencies they oversee.[Footnote 2] FTA 
used the SSO Rail Accident Database to produce the agency's 2009 Rail 
Safety Statistics Report, which analyzed data from 2003 through 2008. 
[Footnote 3] Although we originally intended to report on safety 
trends using the SSO Rail Accident Database in the rail transit 
report, we determined that these data were not sufficiently reliable 
for such a purpose. As a result, in this review we further assess the 
SSO Rail Accident Database and FTA's processes for collecting and 
compiling the data. We determined that there are numerous inaccuracies 
in FTA's SSO Rail Accident Database and, consequently, the 2009 Rail 
Safety Statistics Report. FTA implemented changes to the data 
collection process over the past few years to address some of these 
issues. For example, FTA now requires SSO agencies to report incident 
specific information instead of aggregated or summary totals. In 
addition, FTA revised the definition of rail grade crossings to ensure 
consistent accident reporting by rail transit agencies, and now 
requests SSO agencies to provide unique incident tracking numbers to 
assist with data validation and prevention of duplicate entries. FTA 
is also working to validate and correct existing data in the SSO Rail 
Accident Database. Despite these changes, the SSO Rail Accident 
Database and the 2009 Rail Safety Statistics Report still include 
erroneous data. 

DOT is seeking legislative authority to directly regulate and enforce 
rail transit safety, in part on the basis of recent trends in 
accidents and injuries. There were several legislative proposals 
introduced during the 111th Congress to give FTA authority to 
establish safety regulations for rail transit agencies and, in 
cooperation with the states, oversee and enforce compliance by these 
rail transit agencies with these regulations.[Footnote 4] If FTA 
assumes greater oversight authority over rail transit agencies, 
accurate and reliable rail safety data will be critical. The purpose 
of this report is to convey our findings about how insufficient 
internal control over the entry and processing of rail transit safety 
data has negatively affected the reliability of the data contained in 
FTA's SSO Rail Accident Database and its 2009 Rail Safety Statistics 
Report. 

Summary of Results: 

Our analysis of the reliability of data contained in the SSO Rail 
Accident Database found data discrepancies, such as unverified and 
duplicative entries, as well as insufficient internal control. 
Consequently, FTA's 2009 Rail Safety Statistics Report, which focuses 
attention on safety issues in the public rail transit industry, and is 
intended to support FTA's mission to identify safety priorities and 
strategies to address industry concerns, includes unreliable data. 
This affects the accuracy of the statistics regarding rail transit 
safety, including those used for producing information on trends over 
time, which can be safety indicators to help guide FTA's safety 
oversight efforts. For example, data discrepancies may under or 
overstate accidents, injuries, fatalities, and property damage totals, 
among other things. FTA has implemented some changes to the data 
collection process and has begun to validate data and correct 
discrepancies contained in the SSO Rail Accident Database. However, 
FTA's validation proposal does not contain specific efforts to 
establish procedures that would improve data reporting in the future. 
We are making recommendations to DOT to help FTA improve internal 
control over the process used to obtain data from SSO agencies and 
ensure the accuracy and reliability of the SSO Rail Accident Database. 

Scope and Methodology: 

The information in this report is based on work conducted from 
September 2010 to January 2011, in accordance with generally accepted 
government auditing standards. Those standards require that we plan 
and perform the audit to obtain sufficient, appropriate evidence to 
provide a reasonable basis for our findings and conclusions based on 
our audit objectives. We believe that the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. In conducting our work, we interviewed FTA officials and 
analyzed SSO and rail transit agency data to assess the extent to 
which rail transit safety data used for FTA's rail safety statistics 
report were complete and accurate and avoided unverified and omitted 
data entries, duplicative entries, and other data discrepancies. For 
our engagement, we focused primarily on the SSO templates and FTA's 
SSO Rail Accident Database since these are used by FTA to produce the 
rail safety statistics report. In order to ensure we were assessing 
the exact data FTA uses, we replicated the first five summary tables 
presented in that report. We reviewed SSO data from 2003 through 2008 
and assessed the extent to which FTA reconciles the data reported by 
SSO and rail transit agencies with the National Transit Database (NTD) 
and other supplemental sources such as the National Transportation 
Safety Board (NTSB), and how the FTA's procedures contribute to the 
reliability of the data in its rail safety statistics report. To 
determine the extent to which FTA's internal control over the entry 
and compiling of rail transit safety data influences the reliability 
of the data and annual reported statistics, we reviewed FTA's rail 
transit safety data collection policies and procedures for SSO 
agencies and rail transit agencies and identified examples of data 
discrepancies such as unverified and duplicate entries. We assessed 
FTA's processes against GAO's Standards for Internal Control in the 
Federal Government. 

Background: 

FTA relies on two primary information sources for its SSO Rail 
Accident Database: SSO agency data and the NTD. SSO agencies are 
required annually to submit data to FTA on rail transit accident and 
hazard investigations including date, type of accident, number of 
injuries, number of fatalities, probable cause, property damages, and 
type of individuals injured, such as passenger or worker. These annual 
data submissions are manually entered into a template that the SSO 
agencies submit to FTA. FTA recommends, but does not require, SSO 
agencies to provide internal tracking numbers assigned to each 
accident, but not all SSO agencies do so. For example, in 2008, SSO 
agencies for North Carolina, Illinois, Louisiana, and Utah did not use 
internal tracking numbers to record their accidents. The NTD is a 
separate system that includes information on rail safety reporting, as 
well as operating statistics. Transit agencies are required to provide 
data monthly to the NTD through an Internet-based reporting system. 
NTD data include more detailed information compared with SSO agency 
reports, such as the incident time, location, and descriptions, but it 
does not contain probable cause determinations. NTD data also have an 
incident number unique to NTD that does not correspond to SSO 
agencies' tracking numbers. According to FTA officials, FTA's Office 
of Safety and Security relies heavily on contractors to compile 
information primarily from SSO agencies and the NTD system into the 
SSO Rail Accident Database. This information may be supplemented with 
data from the National Highway Traffic Safety Administration (NHTSA), 
the Federal Railroad Administration (FRA), the NTSB, and the National 
Safety Council (NSC). FTA used the SSO Rail Accident Database to 
produce the agency's 2009 Rail Safety Statistics Report, the most 
current safety report available. 

FTA's Lack of Internal Control Contributes to Unreliable Rail Transit 
Safety Data: 

The SSO Rail Accident Database contains data discrepancies such as 
unverified and duplicative entries, rendering the data unreliable. 
Although we have not fully assessed the extent of reporting errors, we 
have found sufficient problems with internal control to warrant our 
concern about the reliability of FTA's SSO Rail Accident Database and 
reports produced from these data. According to GAO's standards for 
internal control in the federal government, internal control is a 
major part of managing an organization and comprises the plans, 
methods, and procedures used to meet missions, goals, and objectives. 
Internal control supports performance-based management and also serves 
as the first line of defense in safeguarding assets and preventing and 
detecting errors and fraud. In short, internal control, which is 
synonymous with management control, helps government program managers 
achieve desired results through effective stewardship of public 
resources. Control activities specific for information systems that 
would help ensure completeness and accuracy of FTA's SSO Rail Accident 
Database include (1) all authorized transactions entered into and 
processed by the computer; (2) reconciliations performed to verify 
data completeness; (3) the agency's data entry design features 
contribute to data accuracy; (4) data validation and editing performed 
to identify erroneous data; and (5) erroneous data captured, reported, 
investigated, and promptly corrected. The following examples 
illustrate instances where a lack of these control activities has led 
to unreliable or inaccurate data contained in FTA's SSO Rail Accident 
Database: 

* Unverified data entries and other discrepancies: Unverified data 
entries can affect data accuracy. We found that FTA's SSO Rail 
Accident Database contains a number of unverified data entries. 
Unverified accident reporting may under or overstate accidents, 
injuries, fatalities, and property damage totals, among other things, 
depending on the nature of the discrepancy. 

- There were 84 entries in FTA's SSO Rail Accident Database for years 
2003, 2004, 2005 that had an accident date of January 1. For example, 
FTA's SSO Rail Accident Database lists 34 incidents occurring on 
January 1, 2003, for Maryland Transit Administration (MTA). FTA 
officials stated that these entries occurred because SSO agency data 
was collected and certified in aggregate (i.e., SSO agencies only 
provided summary accident totals) for the year. Beginning in 2006, FTA 
required SSO agencies to provide incident specific information that 
could be more easily validated for accuracy. However, the use of 
January 1 for multiple accidents that did not actually occur on that 
date makes it impossible to verify whether these accidents were 
appropriately included in the rail transit safety report's statistical 
and trend analyses for 2003, 2004, and 2005. 

- Of the 3,666 entries for 2003 through 2008 in FTA's SSO Rail 
Accident Database, 818 (22 percent) were entries from the Southeastern 
Pennsylvania Transportation Authority (SEPTA), the fourth largest 
light rail transit system in the country. This was more than nine 
times the 87 entries for Massachusetts Bay Transportation Authority, 
the fifth largest light rail transit system in the country. FTA 
officials told us that SEPTA reported rail grade crossing accidents 
inconsistently with other rail transit agencies from 2003 through 
2005. For example, SEPTA was including collisions occurring at 
intersections of driveways and alleyways and other midblock locations. 
The effect of this inconsistent reporting is that the entire number of 
reportable accidents included in the SSO Rail Accident Database from 
2003 through 2005 is overstated, which distorts accident rate trends 
in the 2009 Rail Safety Statistics Report. Once FTA identified this 
issue, the agency tightened its reporting definition and provided 
technical assistance on the new definition to obtain more consistency 
in how incidents are reported. As a result, SEPTA began reporting 
accidents in a more consistent manner from 2006 forward. To address 
the inconsistency of rail grade crossing collisions between 2003 and 
2005, FTA is working with SEPTA to review the data. 

- Of the 3,666 accident entries from 2003 through 2008 in FTA's SSO 
Rail Accident Database, 82 percent did not have a tracking number. 
Between 2003 and 2005, agencies did not provide tracking numbers 
because, as discussed above, data were reported in aggregate during 
that time period. We observed that from 2006 through 2008, many SSO 
agencies included tracking numbers in their accident reports to FTA 
during those years. However, FTA's SSO Rail Accident Database did not 
contain any tracking numbers prior to 2008 even though these data were 
submitted to the agency and could be used for internal control. The 
use of unique tracking numbers as assigned by each SSO agency would 
assist FTA in verifying the rail safety data and prevent duplication 
of entries. 

- The SSO Rail Accident Database indicates that an accident occurred 
on the New York City Transit (NYCT) system on April 29, 2007. 
According to the database, there was one worker fatality, no other 
workers injured, and five passengers were injured. FTA officials 
stated that they verified the worker fatality and added the five 
injured passengers from the NTD report of this accident. However, 
according to a NYCT senior official, there was in fact one worker 
killed, one worker injured, but no passengers reported injured in the 
accident. Thus FTA's SSO Rail Accident Database may not include one 
worker injured and may overstate five passenger injuries. 

- SSO accident thresholds require that accidents must be reported if 
they result in "injuries requiring immediate medical attention away 
from the scene for two or more individuals." Below are two examples of 
accidents in the SSO Rail Accident Database that reflect inaccurate 
reporting. FTA officials maintain these are accurately reported 
injuries. We disagree. The inflated quantity of injuries from these 
two examples remain in the SSO Rail Accident Database and were used to 
compile statistics in the 2009 Rail Safety Statistics Report. 

-- A Chicago Transit Authority (CTA) train, number 220, derailed 
between Clark/Lake and Grand/Milwaukee Stations in Chicago, Illinois. 
FTA's SSO Rail Accident Database lists a date of July 10, 2006, for 
this accident, with injuries to 257 passengers and four workers. NTSB, 
as mentioned above, is a source FTA uses to verify data. However, the 
NTSB report for this accident states that "152 persons were treated 
and transported from the scene, including three injured firefighters 
and 1 injured CTA supervisor." In addition, the NTSB report has an 
accident date of July 11, 2006. FTA officials maintain their injury 
count is correct stating that NTSB reports are often not finalized 
until a year or two after the accident. FTA officials also maintained 
that the additional 105 injuries occurred to passengers who claimed 
they received immediate medical attention away from the scene in the 
days following the accident. FTA officials acknowledge that CTA has 
identified approximately 40 of the 105 people who claimed immediate 
injuries in the derailment were not on the train; nevertheless these 
injuries are still included in the database. 

-- In the SSO Rail Accident Database, NYCT reported 1 track worker 
fatality and 19 injured track workers resulting from a collision 
occurring on April 24, 2007. However, the official accident report of 
this accident states that 1 track worker was killed during the 
collision and does not mention any injured workers. Upon contacting 
NYCT, we were able to confirm with the agency that, while they 
reported 19 injured workers, these workers were neither injured at the 
scene nor immediately transported for medical attention, but rather 
suffered trauma from witnessing the accident in the days following. 
This error is significant and resulted in overstating overall injury 
totals, as well as injured worker totals, in the 2009 Rail Safety 
Statistics Report by 19, or about 12 percent of all injured workers in 
the rail safety database for 2003 through 2008. 

* Duplicative entries: Duplications in data entry overstate injuries 
and property damage totals, among other things. In our review of SSO 
agency data, our analysis found 76 of 3,666 records (or about 2 
percent) that were duplicative entries (i.e., the incident was entered 
twice). 

- FTA agreed that 60 of these duplicative entries were a result of 
three agencies submitting duplicate incident data reports to FTA. 
These errors were identified by FTA in 2009, and changes were made 
that same year to the incident update submission process to address 
this flaw. However, FTA used the inflated quantity of duplicate 
accidents from these three agencies when compiling statistics for the 
2009 Rail Safety Statistics Report. 

-- FTA officials stated that about 25 incidents in FTA's SSO Rail 
Accident Database for the calendar year 2008 were entered twice--once 
under "Denver RTD," or Regional Transportation District, and again 
under "RTD." FTA confirmed this error was not caught until after the 
publication of the 2009 Rail Safety Statistics Report. Each duplicate 
incident, and all related data associated with that duplicate record, 
inflates accident rates and trends. For example, the overstatement of 
property damage from this 2008 error was $251,249, and passenger 
injuries were overstated by six. 

-- In February and March of 2007, New Jersey's Department of 
Transportation reported multiple incidents on the same day for the 
Port Authority Transit Corporation (PATCO). FTA officials agreed these 
entries for PATCO were entered twice. 

-- In July 2007, Washington State's Department of Transportation 
reported multiple incidents on the same day for Sound Transit-Tacoma 
Link. FTA officials agreed these entries for Sound Transit-Tacoma Link 
were entered twice. 

- We found 16 additional examples of duplicate entries containing 
identical or near identical information. FTA stated that these 16 
entries to the database were reported by the transit agencies as 
separate incidents, and therefore, FTA does not consider them 
duplicate entries. We disagree and believe these records are 
erroneously included twice in the SSO Rail Accident Database, thus 
inflating statistics in the 2009 Rail Safety Statistics Report. 
Examples of these include: 

-- Two SEPTA entries for September 15, 2004, that include an injury 
are in the database. When we compared the descriptions, one entry 
reported that "an 81-year-old female pedestrian was struck by an 
eastbound Route 36 trolley at Island Ave. and Lindbergh Blvd." and 
that "two shuttle buses" were subsequently requested for stranded 
passengers. The second entry states "an elderly female walked in front 
of trolley" and "two shuttle buses" were requested. 

-- Two entries reporting an injury on the Massachusetts Bay 
Transportation Authority (MBTA) are in the database on September 30, 
2005. FTA officials stated that they appear to be two perhaps related, 
but separate, events with different railcars impounded, and that MBTA 
reported the two incidents separately. However, when we compared the 
descriptions of these two entries, both reported an "unidentified 
white male with lacerations" and referenced the same impound number. 

-- Two entries are in the database reporting a pedestrian struck by a 
San Diego Trolley train, resulting in a fatality on May 2, 2005. FTA 
officials stated that two separate reports were submitted for this 
event, and it was not a duplicate entry. However, the location, car 
number involved, and property damage total contained in these two 
records are all identical. In addition, the incident description is 
nearly identical describing a male pedestrian on an "elevated concrete 
structure" who lost his footing and fell into the path of the train. 

FTA has identified the cause of some of the data inaccuracies and made 
some changes to reporting requirements, such as defining rail grade 
crossings more consistently, and requiring SSO agencies to report 
incident specific information instead of aggregated or summary totals. 
However, FTA needs to establish appropriate information system control 
activities, such as those mentioned earlier in this report, in order 
to improve data accuracy and completeness in any future data 
collection efforts. For example, if FTA's SSO agency reporting process 
included data entry design features that prevented unverified or 
duplicative entry and required certain fields to be completed, such as 
incident tracking numbers, locations, and descriptions, some of the 
data discrepancies listed above would likely not have occurred. FTA 
officials have acknowledged the important role that data play in 
making decisions to address challenges to rail transit safety. 
However, our analysis of the data reliability concerns we have 
identified--including data discrepancies such as unverified and 
duplicative entries--along with insufficient internal control, 
demonstrates the unreliability of FTA's rail transit safety data. 
Moreover, the SSO Rail Accident Database and the 2009 Rail Safety 
Statistics Report still include erroneous data. 

The lack of reliable data limits FTA's ability to produce accurate 
accident rates and trend information. As mentioned above, FTA 
officials acknowledged data inaccuracies and inconsistencies and have 
implemented changes to the data collection process over the past few 
years to address some of these issues. In addition, FTA is currently 
working to validate the SSO Rail Accident Database by conducting data 
comparisons with NTD and contacting rail transit agencies to identify 
and correct for discrepancies, as appropriate. However, the validation 
effort proposed by FTA to correct inaccuracies for previous years does 
not contain specific efforts to establish procedures that would 
improve data reporting in the future, such as internal control over 
SSO agency reporting templates that are submitted to FTA, to ensure 
completeness and accuracy. More complete and accurate data, and 
reporting by SSO agencies, would better enable FTA to identify safety 
priorities for rail transit, determine the effectiveness of its 
programs, and provide an accurate picture of overall transit safety. 

Recommendations for Executive Action: 

We recommend that the Secretary of Transportation direct the 
Administrator of FTA to take the following two actions: 

* Develop and implement appropriate internal control activities to 
ensure that the data entered into SSO agency reporting templates are 
accurate. To accomplish this, the Administrator should consider data 
entry design features to ensure consistency in reporting across rail 
transit agencies. 

* Incorporate appropriate internal control over the method used to 
review and reconcile SSO agency data with other data sources to better 
ensure accuracy and reliability of the SSO Rail Accident Database. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to the Department of Transportation 
for comment. DOT agreed to consider the recommendations in this report 
and provided technical comments and clarifications, which we 
incorporated, as appropriate. 

We are sending copies of this report to the Chairs and Ranking Members 
of the Senate Governmental Affairs Committee and the House Government 
Reform Committee and other appropriate congressional committees, as 
well as to the Secretary of Transportation and the Administrator of 
the Federal Transit Administration. The report also is available at no 
charge on the GAO Web site at [hyperlink, http://www.gao.gov]. 

If you or your staff members have any questions concerning this 
report, please contact me at (202) 512-2834 or wised@gao.gov. Contact 
points for our Offices of Congressional Relations and Public Affairs 
may be found on the last page of this report. Major contributors to 
this report were Brandon Haller, Assistant Director; Maren McAvoy; 
Martha Chow; Antoine Clark; Kathleen Gilhooly; Judy Guilliams-Tapia; 
Hannah Laufe; Grant Mallie; and Anna Maria Ortiz. 

Sincerely yours, 

Signed by: 

David J. Wise:
Director, Physical Infrastructure Issues: 

[End of section] 

Footnotes: 

[1] GAO, Rail Transit: FTA Programs are Helping to Address Transit 
Agencies' Safety Challenges, but Improved Performance Goals and 
Measures Could Better Focus Efforts, [hyperlink, 
http://www.gao.gov/products/GAO-11-199] (Washington, D.C.: Jan. 31, 
2011). 

[2] Through the State Safety Oversight Program, FTA monitors 27 state 
agencies that oversee the safety and security of rail fixed guideway 
systems in 26 states. See 49 C.F.R. part 659. One state, Illinois, has 
two oversight agencies, each overseeing a different rail transit 
agency. 

[3] DOT, FTA, 2009 Rail Safety Statistics Report (Washington, D.C: 
2010). 

[4] Public Transportation Safety Act of 2010, S. 3638, 111th Cong. 
(2010); Public Transportation Safety Program Act of 2010, S. 3015 and 
H.R. 4643, 111th Cong. (2010). 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: