Kim, John; Mattson, LeRoy; Sahni, Ashweni (1991, ASQC) Medtronic, Inc., Minneapolis, MN
Because implantable medical devices are life supporting or life sustaining devices, reliability is a critical issue. Devices such as cardiac pacemakers, heart valves, and drug pumps have a long service life, measured in years.
Obtaining data concerning reliability of implanted devices requires the participation of several participating clinical centers over a considerable length of time. Since actively following these implanted devices can be very time consuming, expensive and often impossible, manufacturers often rely on passive systems of data collection. Passive data collection, however, often lacks complete and accurate data.Various laboratory testing methods and the collection of field data help provide reliability estimates; however, field data is often preferred because of the difficulties involved in simulating long-term human implant condition through laboratory testing. The proposed method presented in this paper is an extension of the life table procedure known as the Cutler-Ederer Method. The life table is an easy-to-use tool that provides a visual representation of the device's performance in the field.
This methods presented in this paper have successfully adjusted for underreported mortality by: (1)mathematical derivations involving the parameters for life table data, and (2) a simple numerical example. These methods also have had a significant impact on adjusting the survival probabilities and number of actives in a real application.
Statistics,Health care,Medical devices,Life testing,Reliability,Biomedical