Articles

The Virtue of Failure

Published on July 1, 2011 by Tony Wickman in Tony Wickman, CTPO

Not all of my ideas are good ones, and neither are yours, but you know what? That’s not necessarily a bad thing. Frequently I am asked to build an orthosis, which, in my humble opinion, is bound to fail. It could be because of the design, the materials chosen, or even the components, but for some reason it just doesn’t seem like it will work. Since my job first and foremost is to build braces, I have to yield to the person who knows the patient’s needs the best—the practitioner. My most important role, however, is as a technical liaison.

Our roles as technicians are, theoretically, pretty well defined. The practitioner has the clinical skills to understand each patient’s individual pathology and the experience to know what will best suit his or her lifestyle. The practitioner meets the end users, interviews them, reads their medical history, speaks with the referring physicians or therapists, and through all this investigation, comes to have a pretty good idea of the patients’ needs. Once the practitioner has a good idea of the parameters involved and the anticipated outcome, he or she generates a concise plan and then passes that information along to the technician. Then the technician has to take this data and overlay his or her understanding of the materials, designs, and components required to safely achieve the stated goals. Between the two of us—practitioner and technician—we are supposed to be able to work whatever magic needs to happen to help end users meet their orthotic needs. Most of the time this works pretty well, but sometimes the practitioner wants to do better, or the end user wants to do more, or the provider wants to do less, and we fail.

Over-Engineering vs Under-Engineering
Technicians frequently over-engineer a device. For years I was able to brag that I had never had a single prosthetic component I manufactured fail. Then I realized I had actually just admitted I had over-engineered the vast majority of what I had built. We are all guilty of developing a certain level of “scar tissue” in our design rationale. We have had similar devices, on similar patients, fail as a result of one parameter or another, and after a while we tend to play it safe. The problem is, of course, that when we play it safe, we can easily build a device that is substantially more stable than the circumstances require. This causes the device to be heavier or more restrictive than the patient demands. Then it gets put in the closet and is never used. Yes, the device may have worked, but in this situation we failed just the same.

Other times technicians under-engineer a part. This can be even more dangerous because the catastrophic failure of a part can expose the whole team to potential liability. Let’s face it, when people rely on the things we make to ambulate, the device’s failure can lead to serious injury! As much as we want to avoid this, we have to be aware of this balance. Even though it may seem necessary at times, we cannot let the potential for failure stop us from designing appropriate devices. If anything, we need to continue to explore newer, better designs and materials in an effort to move the industry forward.

The first trick to mitigating the risk of failure is to have a good understanding of all the materials we use. Whether we are dealing with metal, leather, plastic, or composites, we have to know the material’s limitations. Since most of the orthoses we make are essentially prototypes, it is very hard to use science to predict the outcome, so we have to rely on an innate understanding of how these materials will react in certain situations. With most materials we need to know the direction of the forces to be applied, the strain, tension, and shear forces they will have to deal with, and the environmental conditions they will operate in. We have to know which materials are appropriate for each design, and the only reason we know these things is because of the failures we have encountered.

Failure Analysis
The only way we can learn from the mistakes of the past is to communicate with each other. Technicians have to have a thorough understanding of all the parameters involved. How many times have you used “B”-sized joints on a 120-pound post-polio patient only to find the person in question is highly active and has a history of breaking joints? This is information practitioners frequently omit from the orthometry form, so as technicians, we have to be careful about interviewing practitioners to get any information we may need to make a decision about component selection. Ideally, the practitioner will specify these parameters, but if they do not, you need to be bold about acquiring this data! When a failure does occur, we need to debrief the whole team. Everyone needs to understand the failure and learn from it.

Things break. We know that and we can’t always see it coming. When they do break we need to know why! So failure analysis is critical. We need to know the circumstances involved in the incident, what the user was doing, how old the device is, if the device was under-engineered, and if the device was used beyond its original parameters. There is a science to failure, and each occurrence can tell us a great deal about how and why a design failed. Don’t just throw a broken part away! Stop and examine it! Most devices will give up their secrets pretty easily. Between a basic interview with the user and a visual analysis, you can usually see what went wrong. If the part in question was manufactured off site, it needs to go back to whoever made it; the manufacturer is the one who most needs to understand the failure!

One of the biggest design obstacles this industry faces is the lack of a central database for failures. With the advent of evidence-based practice, we can begin to develop a more fundamental view of the potential for failure. This is good, but if we have several hundred separate pools of “evidence,” the industry will be slow to react to the potential for a given design to fail. If this data could be collected and centralized, it would afford a much more uniform approach to the designs of tomorrow. There’s at least one repository for failed components operating today, but the availability of a U.S. repository with critical data about the failure—not only of components but also of finished devices—would go a long way to establishing some basic design criteria for the products we make and secure a more positive, and safer future for the people we serve.
Earl Nightingale said, “When we succeed, we party, and when we fail, we ponder.” Only through failure can we establish boundaries and build a foundation for the development of a better industry.

Also published in the July 2011 edition of the O&P Edge. © 2011 O&P Edge