Sometime between June 1985 and January 1987, a computerize radiation therapy machine gave some abnormal doses of medication to patients. Therac-25 massively overdosed patient which resulted in severe injuries and in some cases even led to death of the patient. In the medical industry overdoes usually occur due to operator error but in this case is occurred due to the software that the user was operating on the instrument. The manufacturer of this machine was too over confident about their product that they did not do a thorough test of the software. The manufacturer had a lot of confidence in the software’s ability to ensure safety of the patient. An older instrument, Therac-20, used others means such as mechanical interlocks to prevent overdoses to the patient. However, Therac-25 depended solely on software.
The ethical problem in this article was the manufacturer’s lack for patient safety. To avoid this accident the manufacturer should have placed more safety features on the machine. It is understandable that they were confident on their software, but should have placed a mechanical interlock or any other means to prevent overdose to patients. That would be like a second safety feature. All machines should have primary and a secondary safety feature. That way in case the primary fails the secondary safety feature takes over.
Another thing the manufacturer should have done is thoroughly tested their software. The article stated that when the manufacturer did an accident report they did not look at the software. The tried to recreate the error but were not able to so they concluded that it was a hardware error and not software.
An Investigation of the Therac-25 Accidents
http://www.onlineethics.org/CMS/computers/compcases/therac25.aspx
