Running Head: THERAC-25 CASE STUDY

Therac-25 Case Study

Therac-25 Case Study

Therac-25 is a radiation therapy machine that was used for treating patients with cancer. The machine and its predecessors, Therac-6 and Therac-20, was a product from the collaboration of Atomic Energy of Canada Limited (AECL) and a French company called CGR (Leveson, n.d., p. 2). It is this machine that will cause other developers and manufacturers to rethink how to develop machines that will impact people’s lives. The Therac-25 was involved in six massive overdoses that cost three people’s lives, while the machine gave the operator error messages or a message that states that no dose was given (Baase, 2008, p. 425). After the first accident, some users became skeptical of the safety of the device. Yet, the manufacturer turned a blind eye on the matter. They responded to the incident that they have fixed the problem. In the following months, five more accidents occurred. In each incident, the manufacturer, the computer programmer, and the clinic/hospital using the Therac-25 have some responsibilities and flaws that need to be addressed.

The Problems

Manufacturer

What would be the first thing that a manufacturer needs to do when they have an incident with one of their products? It should be the recall of that product, in this case the Therac-25. The Therac-25 can help a person ailing with cancer with its beam but the beam itself would be too harmful to the patient. Any medical equipment that has the potential to harm a human being should be carefully used. The manufacturer was confident of how safe the machine was and installed them at different locations. When the manufacturer learned that the first victim had radiation burns, they “refused to believe that it could have been caused by the Therac-25” (Leveson, n.d., p. 11). They believed that the Therac-25 is the state-of-the-art machine that helps patients rid of cancer, not a contraption that harms patients. It is understandable that people become protective when their pride is at stake but it is necessary to investigate an incident when it involves people being harmed. Another issue in overconfidence is when the manufacturer removed the hardware safety device and left safety decisions to the software (Baase, 2008, p. 428). The manufacturer was able to save money by removing parts that was assumed to be not needed. Cotterman, Forsberg, and Mooz stated in their book Visualizing Project Management (2005) that the Therac-25 “was designed to be cheaper to produce and operate” (p. 251). The manufacturer may have removed many parts of the machine to reduce the cost of producing the Therac-25. Is cheaper better? Not necessarily. The accidents that occurred with the Therac-25 explain that fact. The accidents caused by the Therac-25 are not only the manufacturer’s fault. The programmer also has some responsibilities in this matter, which is discussed next.

Programmer

Usually development teams have several programmers to complete a project. Each programmer may work on different parts of the project so that much of the project would be complete before testing begins. The software in the Therac-25 was created by a single programmer (System Safety, n.d., para. 3). What can one programmer do in a limited amount of time? Rather, did the programmer have the proper credentials to create software? Unfortunately, before any information about the programmer was collected, they left the manufacturer and their whereabouts are unknown (Rose, 1994, para. 46). Although, no information about the programmer exists, their credentials are questionable. To create proper software, programmers are required to document everything that they do to the software. Even changing a single character in a code is required to be documented. It is so that if the software causes bugs to appear, which is frequent in programming, the programmers can check the documentation to see what they did at a certain point in the project and try to fix it. The manufacturer, rather the programmer, created little documentation on the Therac-25 (Baase, 2008, p. 426). The manufacturer may have rushed the programmer to create the software for the Therac-25 but that does not mean the programmer not think twice about documenting what they did. It would have helped if they kept small notes of what they did if they were in a rush. One thing that the programmer should have done was rewrite the code for the Therac-25. What they did was they reused the code from Therac-20, which was from Therac-6, which they thought worked properly (Leveson, n.d., p. 4). It saves time to reuse code but the code that will be reused need to be checked for bugs or else it will be present in the new software. The software used in the Therac-20 had some bugs in it but safety measures prevented it from becoming apparent until the Therac-25 produced similar errors as the Therac-20 (Cotterman et al., 2005, p. 251). It may have helped if new code was written and tested extensively to prevent the Therac-25 accidents from happening.

Clinics and Hospitals

The clinics and hospitals that used the Therac-25 successfully treated many patients until the accidents occurred. The staff ignored some of the responsibilities that they have when these accidents happened. They may have been confused or did not know how to respond to the Therac-25 accidents rather than being ignorant but some can be quite ignorant of other’s needs. One doctor at a facility where the first accident took place stated that they did not know how to respond to the matter of the overdose because they never seen such devastating effects from the Therac-25 (Leveson, n.d., p. 10). Overconfidence plays a role in the staff’s ability to reason with the patient. A staff was confident that the Therac-25 did not burn a patient, even though the patient was complaining that it did (Baase, 2008, p. 428). The staff should have taken the patient’s complaints and tried to accommodate with the patient’s needs. There are other things that the staff ignored that was life threatening to the patient. At one facility, a staff repeatedly resumed a treatment because numerous error messages disrupted the procedure, while the patient was overdosed with radiation (Leveson, n.d., p. 12). Ignorance can be a very deadly trait. Clinic and hospital staff should be able to immediately accommodate with the patient’s needs.

Conclusion

The Therac-25 had a history of costly experiences. It successfully treated many patients before the accidents began. It is the clinic and hospital staff that found many flaws in the system, yet they were ignorant of the patient’s needs when the patients were feeling pain after the treatment. Each entity involved in the accidents should have kept in mind that the safety of the patients was the highest priority. This incident should be studied more so that the same problem will not happen again.

References

Baase, S. (2008). A Gift of Fire: Social, Legal, and Ethical Issues for Computing and the Internet. (3 ed.). Upper Saddle River, NJ: Pearson Prentice Hall.

Cotterman, H., Forsberg, K., & Mooz, H. (2005). Visualizing Project Management. (3rd ed.). New York: John Wiley & Sons.

Leveson, N. (n.d.). Medical Devices: The Therac-25. Nancy Leveson's Home Page at MIT. Retrieved July 26, 2010, from http://sunnyday.mit.edu/papers/therac.pdf

Rose, B. (1994, June). Fatal Dose - Radiation Deaths linked to AECL Computer Errors. The Canadian Coalition for Nuclear Responsibility. Retrieved July 26, 2010, from http://www.ccnr.org/fatal_dose.html

System Safety. (n.d.). Welcome to ComputingCases.org. Retrieved July 26, 2010, from http://computingcases.org/case_materials/therac/analysis/Safety.html