Computing Cases Header, Picture of a Keyboard with the text "" printed over it

Case Materials

Case Navigation

Therac-25 Case

Teaching Intro

Socio-Technical Analysis

Ethical Analysis


Supporting Documents



Teaching Tools

Teaching with Cases

Social Impact Analysis

Computer Ethics Curriculum

Curricula Index

Case Materials



Hughes Aircraft

Ethics in Computing Links

Contact Us

Summary of Operator Interview

The following article is the result of an interview we conducted with a Registered Therapy Technologist who has extensive experience operating medical linear accelerators. This individual currently manages a Radiation Therapy Department at a major United States hospital, and trains technicians to operate radiation therapy machinery. For privacy purposes, the true identity of this person will remain anonymous, and for the remainder of the article, we will refer to our interviewee as "Susan."

Susan operated a Therac-4 linear accelerator machine in the mid 1980's. At the time, Susan had recently graduated and was working at a University where the radiation therapy technology was fairly advanced. She enjoyed operating AECL's Therac machine because it was one of the first computerized linear accelerators. Looking back, Susan remembered that while operating the machines, she did not think much about whether there could be computer software "bugs" in the system. The technology was new, and she remembered trusting the machine's components and its designers.

When recalling the advantages of the new computerized machine, Susan reported being able to move more patients through during the day. She also remembered feeling good about the extra time she had to talk with patients when she was working with a computerized machine.

Susan learned about the Therac-25 incidents while attending a national radiation therapy conference in 1990. A radiation therapist who was also a lawyer gave a lecture on the Therac-25 accidents. He handed out newspaper articles about the incidents and spoke about how many times the therapists involved in the accidents attempted to resume treatment in spite of the error messages they received from the computer. The lecture focused on the question of how many attempts to resume treatment is too many? The lecturer and the participants discussed the possibility of establishing institutional policies and limits on the number of times an operator could resume treatment after having received an error message, such as the cryptic "malfunction 54" messages that the operator received during the two fatal accidents in Texas.

The problem, Susan reported, is that back in 1990, and today, there are no industry-wide standards or rules for these types of situations. Susan felt that she had been lucky to have always worked where there was a physicist available to provide help with the many error messages operators received. She also felt that in other clinics, where this kind of assistance is not available, there was, and still is, a great deal more pressure on therapists to just keep going despite the error messages. An operator might attempt, for example, to deliver the prescribed dose in 12 increments instead of 1 by continually clearing the faults generated by the computer. Susan stated that this type of activity happens all the time in medical radiation therapy, particularly in clinics where there is more pressure from the administration to keep patients moving through quickly.

Although Susan had been working with a AECL Therac machine at the time of the accidents, she did not remember receiving warning notices from AECL about the Therac-related accidents. Susan believes that this is one aspect of the industry that has changed, possibly, in part due the Therac-25 accidents. At the present time Susan receives notices from the manufacturers of the linear accelerators used at her hospital whenever there is a linear accelerator malfunction, or even if there is a malfunction that almost occurred, but was prevented.

Perhaps part of the reason that Susan did not hear of the Therac-25 incidents until much later was that the hospital where she worked got rid of the Therac-4, moved their facilities, and bought a new set of linear accelerators. Susan estimated the average life of the linear accelerator to be between 5 and 10 years. After that, she said, the accelerator tends to act somewhat like an old car in which the engine light is coming on all the time. The accelerator's computer generates many faults that can become a nuisance to the operators and to the patients. Responsible operators will continue to report these faults to the physicist, when one is available, and eventually, the machine is replaced.

Susan feels that one of the biggest problems in her industry today is the lack of rigorous industry-wide standard certification and education for operators. Susan reported that there are about 102 radiation schools in the country, and that there are also different types of schools. Students are able to receive a certificate from a certificate program, usually about 12 months in length. Students are also able to receive a four-year bachelor's degree from certain schools. The American Registry of Radiologic Technologists (ARRT) provides a test that graduates of these programs may then take in order to be considered licensed entry level technicians. The ARRT also requires that therapists maintain their training through continuing education. Therapists must have 24 credits in two years before they may re-register their licenses.

In spite of the fact that the ARRT provides these guidelines for licensure, many states in the U.S. do not require hospitals or clinics to hire licensed radiation therapists. Some states require very basic exams, but, according to Susan, that in essence means that in many states anyone off the street could learn how to operate a machine, take one of these basic exams, and then be qualified to operate radiation therapy machines.

Susan and many of her colleagues continue to fight for mandatory standard certification of radiation therapists. The safety of patients depends on all of the elements of their systems of treatment working together correctly. The more operators are trained to know about the process, the more they will be able to help prevent accidents. Well-trained operators can double-check radiation dose prescriptions and question doctors when something does not seem right. With the benefit of extensive training, operators have a better sense of when it is alright to over-ride a fault message from the computer.

Well trained technicians will also be better equipped to stand up to hospital administrations that attempt to put pressure on technicians to push large numbers of patients through treatment in spite of possible dangers. Though Susan does not feel this kind of pressure from her own administration, she knows that other technicians in other clinics definitely do, especially at "free-standing" clinics that operate for profit. Susan is aware that at these clinics there is a tremendous amount of pressure put on machine operators to get patients through treatment.

Susan also described incidents in which technicians left institutions because they didn't feel that the institutions' radiation therapy practices were safe for patients. Because there is no federal law regulating how many times an operator can re-attempt therapy after the computer displays a fault or shuts down, some operators allegedly use jumper cables that continuously override their computer's emergency shut down mechanism. Susan cited a lack of regulation, lack of training, and lack of adequate funding as reasons for these procedures.

Another issue in the radiation therapy industry that worries Susan is the fact that linear accelerator manufacturers charge large fees for operator training sessions, software upgrades, and machine maintenance contracts. When a radiation therapy machine is purchased, it comes with many binders full of information provided by the company. The clinic is given the option to buy service contracts and send physicists and operators to the company headquarters for training. Susan reported that in many clinics where money is tight, administrators are forced to choose between machine servicing contracts, software upgrades, and training.

According to Susan, mistakes are still made in the radiation therapy treatment of patients. Much of the information and calibration is still done by human beings and subject to human error. As an instructor, Susan teaches her students to anticipate every angle of the treatment, and then to check, and re-check their work. Susan also mentioned that while she teaches her students not to trust wholly in the machinery and its software, operators are largely dependent on manufacturers and hospital physicist teams to keep the machines running correctly.

Susan has a positive outlook regarding the radiation therapy industry. She knows that thousands of patients benefit greatly from radiation therapy technology. While Susan continues to push for operator certification legislation, she focuses on training her own staff well. Susan and her administration also focus heavily on quality patient care.

When asked if she thought it would be important for the designers of the software that runs the machines to know what it is like to do her job, Susan's reply was an emphatic yes, though she doubted many of the software designers of her machinery had spent much time observing a radiation treatment facility.