The Honorable Jim Hall
National Transportation Safety Board
490 L'Enfant Plaza East, S.W.
Washington, D.C. 20594
Dear Chairman Hall:
This is Greyhound Lines, Inc.'s response to your January 14, 2000 letter concerning the Greyhound bus accident that took place on the Pennsylvania Turnpike near Burnt Cabins, Pennsylvania on June 20, 1998. Your letter sets forth the Board's determination of the probable cause of the accident and its safety recommendations for Greyhound. Although Greyhound strongly disagrees with the Board's probable cause determination, we have carefully reviewed all of the Board's recommendations and have implemented them where appropriate.
This letter is divided into two parts - the first addresses the probable cause issue and the second describes Greyhound's response to each of the Board's recommendations.
Greyhound believes that the cause of the accident was that the accident driver suffered a severe cardiac event that caused him to lose control of the bus. Two forensic pathologists retained by Greyhound independently examined tissue slides from the accident driver. Both found hemmorhaging within the coronary artery and concluded that this led to the coronary event that caused the accident. Dr. Halbert Fillinger stated:
"It is therefore my opinion, with reasonable medical certainty, that the site of the hemorrhage within the coronary artery wall, which produced a reduction in the caliber of that vessel deprived the myocardium of sufficient oxygen with a resultant ischemia and acute anginal distress which then distracted the driver, causing him to strike the back of the parked tractor trailer."
Dr. Emanuel Rubin concluded:
"The most likely scenario in this case was 1) hemorrhage into a coronary plaque, 2) acute myocardial infarction, 3) sudden cardiac arrhythmia and 4) failure of the circulation and loss of consciousness."
Board investigators then consulted an Armed Forces medical examiner, Colonel William Gormley. Although Col. Gormley agreed that the plaque hemmorhage was present, he concluded that "in the absence of documented signs or symptoms of myorcardial ischemia just prior to the collision, this autopsy finding cannot be used to determine the role of the coronary artery disease in the mishap". He explained what he meant by the "absence of documented signs or symptoms" as follows:
"When drivers become incapacitated from arrhythmia associated coronary vascular disease there is a brief period of consciousness after the arrhythmia reduces or eliminates cardiac output. People usually remain conscious for 6 to 10 seconds after total loss of blood flow to the brain. During this brief period, most drivers would at least be aware that something was wrong and their most likely reflex action would be to slow their vehicle". (emphasis added).
Clearly, Col. Gormley was not aware that the Board did, in fact, have evidence of such an attempt to slow the vehicle by the accident driver. At the January 5, 2000 Board meeting, Board staff revealed photographic evidence of 28 feet of tire marks probably caused by the bus just prior to the collision. Had Col. Gormley known that this "documented sign" was present, he likely would have reached the same conclusion as Drs. Fillinger and Rubin concerning the cause of the accident. Also, the fact is that the driver steered the bus into an emergency pull-off, which is also consistent with the driver being aware that something was wrong and trying to adjust to it.
In sum, the record now shows that there are two expert witnesses who concluded that the driver had a severe cardiac event, which caused the accident. A third expert witness agreed that the necessary circumstances for a cardiac event existed at the time of the accident and strongly implied that he would also find that such an event occurred if there were evidence that the driver tried to slow the vehicle. Since we now know that such evidence exists, the statements of all three expert witnesses support the conclusion that the driver suffered a severe cardiac event, which caused the accident.
By comparison, there is virtually no evidence supporting the conclusion that the accident was caused by the driver's drowsiness brought on by his taking a Benadryl and having an irregular work schedule. The following points demonstrate that lack of evidence:
There were no witnesses that suggested the driver was having trouble with drowsiness while driving that night.
Fellow drivers who talked with the accident driver in Harrisburg roughly an hour prior to the accident found him to be normal and excited about his last trip before retirement.
The accident driver was making his last trip prior to his retirement and had his wife and son with him; it is highly unlikely that he would be drowsy given the excitement of this special occasion.
The behavior observed by the state trooper is just as explainable by the onset of a cardiac event as by drowsiness.
The only evidence of prior drowsy driving was the uncorroborated statement of a witness who claims to have seen drowsy driving by the driver six months prior to the accident; however, this witness' credibility is in question since she admitted that she continued taking trips with this driver; finally had an argument with him and was ejected from the bus; and failed to file a complaint with the company despite being given the procedures to do so.
The accident driver worked a highly regular shift involving 4 days on and 2 days off and had worked this shift for an extended period of time. It involved the same route and the same schedules and although the schedule departure times varied somewhat, the driver basically drove at night and slept during the day. Greyhound's statistics indicate that its nighttime driving accident rate is roughly half what it is during the daytime.
The accident driver had ample opportunity for quality sleep prior to the accident. The record shows that in the 3 days prior to the accident, the driver had checked into a hotel prior to driving for periods ranging from more than 8.5 hours to more than 10 hours.
During the visit to his personal physician seven months before the accident when the driver complained about difficulty sleeping, the driver also complained about chest pains.
Finally, it is important to note that the Board's sole expert support for its conclusion that an irregular driving schedule was the probable cause of this action, is the Board's 1995 study entitled Factors that Affect Fatigue in Heavy Trucks. But to the extent that the Heavy Truck Study is relevant at all, it undercuts the Board's conclusion herein.
The Heavy Truck Study analyzed factors in heavy truck accidents including "irregular" work schedules as defined by the Board. The Report did not analyze any bus accidents, nor any bus accidents involving regular schedules with small start time variations such as those operated by the Burnt Cabins driver.
Moreover, the Board in the Heavy Truck Study concluded that it "could not determine if irregularity of duty/sleep patterns, per se, would lead to fatigue; the drivers with irregular patterns in the Board's study did not receive sufficient sleep in their most recent sleep period" (Heavy Truck Study at 34). The Board also found that "it is noteworthy in this unique sample of actual accidents that factors that affect the ability to obtain adequate sleep, such as irregular duty/sleep and inverted schedules (which are often assumed to be closely associated with fatigue), ranked well below the factors that affect the quantity and quality of sleep". (Heavy Truck Study at 37). In this case, the record clearly shows that the accident driver had ample sleep periods in the several days prior to the accident.
Given the strong expert evidence that the driver suffered a severe cardiac event and the lack of any substantial evidence of drowsy driving at the time of the accident, Greyhound believes that the probable cause of the accident was a severe cardiac event causing the driver to lose control of the vehicle.
Responses to the Board's Recommendations
The following are Greyhound's responses to each of the four Board recommendations to Greyhound.
Revise your driver scheduling practices to reduce scheduling variability that results in irregular work-rest cycles.
As previously indicated, Greyhound does not believe that the regularity, or lack thereof, in the accident driver's work schedule had any causal relationship to the accident. Moreover, Greyhound drivers generally have very regular work schedules, and Greyhound has an ongoing mechanism for revising schedules, in conjunction with the drivers and their union, in order to continually improve safety and driver quality of life.
Periodically, Greyhound drivers bid on the runs that they want to operate. Once their runs are assigned, they operate that same run until the next bid. That means that every day that they work, they operate at approximately the same time of day or night and over the same route. This enables them to establish a work/rest routine that gives them ample rest and maximizes their quality of life. This contrasts sharply with the irregular driver operations of many trucking companies and bus charter and tour firms.
Four times a year, Greyhound adjusts its bus schedules. After each one of those schedule changes, Greyhound, in conjunction with its drivers and the union, adjusts the driver bids in light of the new schedules. The primary factors considered in making the bid adjustments are safety and driver quality of life. This ongoing cooperative effort produces run bids that maximize the regularity of operations.
Greyhound emphasizes that all of its regular runs are structured to comply with the Department of Transportation's hours of service rules. In fact, we exceed those requirements. For example, while DOT requires 8 hours off between on-duty periods, Greyhound requires its drivers to have 9 hours off when away from home and 10 hours off when at home.
Greyhound drivers have a much more regular driving pattern than the drivers of most truck and bus companies. If the Board believes that commercial motor vehicle operators should have more regularized start and finish times, it would be most appropriate to direct those recommendations to DOT so that they can be thoroughly analyzed and tested and appropriate industry-wide rules adopted.
Include in your drivers' assessment programs all driver traffic and logbook violations.
Although speeding was not a factor in this accident, Greyhound is concerned about the results of the speed tests that the Board performed. Since we were notified of those tests, we have taken a number of actions to enhance Greyhound's speeding oversight program. Those actions include:
- Installation of speed radars in all company safety vehicles. Speed observations are forwarded to driver management for action.
- Increasing management road observations - All District Managers and Area Sales Managers have been tasked with a minimum monthly quota. These reports will be sent to Safety for tracking.
- Addressing speeding as a regular topic in safety meetings nationwide
Greyhound also intends to broaden its assessment program to include driver traffic and logbook violations as recommended by the Board. We are taking the following actions:
- Log-book violations - Greyhound audits 100% of driver logs for hours of service violations. We are now t racking "excessive" speed discrepancies identified through Hours of Service log audits and addressing them with our drivers. Our analysis indicates these are often administrative errors (incorrect mileage on the log) made by extra-board drivers unfamiliar with the exact mileage of the runs. Further it is to be noted that Greyhound's compensation is based on time (hourly pay) and not on mileage. Violations while operating a commercial vehicle - No change to our process. These have always been included in driver assessments.
- Violations in non-commercial vehicles - We are now using them as "red flags" for direct management observation (road check) of commercial vehicle operation and discipline based on that observation. We are assessing the extent to which the law allows discipline to be taken based solely on off-duty violations.
Use all current and future data monitoring and storage capabilities of electronic control modules, electronic control units, and similar technologies to enhance vehicle and driver oversight programs by engaging the specific capabilities of each individual unit's programmed or programmable functions to collect and monitor data including, but not limited to, vehicle speed, revolutions-per-minute, hard-brake or sudden decelerations, and other parameters of vehicle and engine operations.
Greyhound believes that the best way for it to control speeding is through continued implementation of the program described above. Although use of ECM data may prove useful in the long run, issues of reliability, practicality, due process and privacy need to be addressed on an industry-wide basis by the Department of Transportation. DOT also needs to analyze its experience with the rules for usage of electronic data in monitoring pilot performance and apply that experience to commercial motor vehicles. Greyhound is fully prepared to participate in a DOT proceeding to resolve these issues and establish rules for ECM usage, but we are not prepared to implement an ECM driver surveillance program on our own.
Greyhound was a pioneer in the use of electronic devices to enhance safety and monitor driver performance. In the early 1990s, we installed the VORAD radar detection system on our buses. That system proved to be unreliable and caused strong adverse reactions from drivers and passengers. Because the system proved counter-productive, Greyhound removed it from its buses. That experience confirms Greyhound's belief that we should not start a new program of using ECM data to monitor driver performance until all of the above issues have been thoroughly reviewed and resolved by DOT, and industry-wide standards adopted.
Revise your 1-800-SAFEBUS program to ensure that all complaints are included in drivers' files and used in drivers' assessments.
Current Greyhound policy is to have driver supervisors interview drivers concerning all non-anonymous complaints and when there is a pattern of such complaints about a particular driver, to monitor the driver's performance and take appropriate disciplinary action based on that monitoring. Greyhound has not utilized anonymous complaints because of their perceived inherent unreliability.
In light of the Board recommendation, we are going to monitor the volume of anonymous complaints received by the SAFEBUS program in a 60 day period to see if those complaints are useful in highlighting problem drivers and/or serious complaints. If so, we will implement a system of monitoring and using anonymous complaints. We also are reviewing the SAFEBUS scripts to determine whether it is appropriate to press harder to get the name and phone number of complainants, thus providing Greyhound with more non-anonymous complaints.
Greyhound shares the Board's commitment to achieving the highest possible level of motor carrier safety. We appreciate the opportunity to provide you with Greyhound's responses to the Board's recommendations.
Executive Vice President and Chief Operating Officer