Sometimes the news headlines involve mysteries. What happened to flight MH370? Where are the missing schoolgirls in Nigeria? What happened to Madeleine McCann? There is a desperate need for answers.

During my early career, I was responsible for troubleshooting problems with diagnostic tests, and wrote the Immunoassay Troubleshooting Guide for The Immunoassay Handbook. Troubleshooting is about understanding a change in situation from “OK” to “not-OK”. The principles of troubleshooting apply to any unexplained event, from a broken radio to a missing plane.

My first investigation, in 1976, involved discovering why a brown liquid extracted from snails was not working in an estriol immunoassay. It seems like a strange ingredient but sulfatase and glucuronidase in the snail extract were supposed to cleave estriol metabolites so the antibody could bind the estriol in blood samples. Usually it worked very well. I discovered high lead and mercury concentrations, which inhibited the enzymes. The supplier had used contaminated snails rejected for human consumption.

To troubleshoot problems you need to carefully define each problem, list the known facts, and brainstorm all the possible causes, for example using fishbone (cause-and-effect) diagrams. Any causes that conflict with the facts can be eliminated immediately. Then, “tests” should be devised to confirm or eliminate the remaining causes. The tests can involve analysis or experimentation. As I learned from bitter experience, jumping to the “most likely” cause without this process usually fails

The following comments are based on media coverage of the investigation into missing Malaysia Airlines flight MH370. I do not have inside information about the actual approach used, so no criticism is intended.

Initial coverage focused on terrorism, then pilot suicide. Both were among the possible explanations but every possible cause needed parallel investigation. Counter-intuitively, focus, which is normally considered a leadership virtue, is not helpful in troubleshooting. In this type of situation a spokesperson must necessarily present a simplified message to distressed relatives and the public, while the troubleshooting team follows multiple leads.

I learned from bitter experience that although experts are essential for their technical knowledge, their opinions about the most probable cause are of surprisingly little value. Recent research has shown that experts are even worse than non-experts at estimating the relative probability of a particular explanation.

Once, I started a 3-day holiday weekend troubleshooting a product failure, working with 3 experts and a laboratory technician. We reviewed the possible causes, suggested by the experts, then the technician pointed out that a humble wash buffer reagent was not on the list. The experts dismissed this as most improbable but it went on the list anyway. The first experiment was complex because it investigated many possible causes but it proved that it was the batch of buffer that was at fault and we enjoyed the rest of the holiday with our families.

The theory that terrorism was a probable cause of the Malaysian Airline plane’s disappearance was reinforced when it was discovered that passengers had forged passports, but then an alternative explanation was found.

Suicide or a terrorist act by one of the pilots became the prevailing theory, and the information about two communications systems being turned off at different times was seen to be strongly supportive. Later, alternative possible explanations for this time difference were discovered, and enquiries failed to find any evidence for premeditated action by either pilot.

Terrorism dropped down the priority list. But just because the passengers and the pilots were eliminated as terrorist suspects, that does not mean that terrorism can be eliminated from the list of possible causes. Perhaps the pilots were victims and not perpetrators.

Facts should include things that did not happen. From what we have heard, none of the passengers or crew sent a text message as they flew across Malaysia, after the plane had changed course 40 minutes into the flight. This is a relevant fact, because it could rule out several scenarios. Investigators tend to focus on facts that fit the prevailing theory, but all facts must be collected until the cause is found. In the early stages, facts should be used to eliminate possible causes, rather than support prevailing theories.

The changes in flight direction away from the correct route occurred at least twice. If a scenario does not explain this it can be eliminated from the list of possible causes.

Hopes for an explanation are fading unless the plane is found. Psychologically, if investigations lurch from one prevailing theory to another, and these lead to dead ends, the end result is despair and frustration, and people give up after a few theories have been explored. There are many possible explanations that should be gradually eliminated by diligent fact checking and experimentation. Every time a possible cause is disproved it should be appreciated as a step in the right direction. This situation is analogous to geology, archaeology and paleontology, where limited facts must be interpreted and explanations found, without any eye-witnesses or contemporary recordings.

Chronology and coincidences matter. When did the course of the missing airplane change? The events coincidental with changing course are critical. That period is not covered by the cockpit voice recorder, which retains the final 2 hours of audio. This is another reason why troubleshooting should continue in the absence of the plane wreckage, which still may not explain the cause.

The Immunoassay Troubleshooting Guide, which was originally based on more than 100 successful investigations, was the first chapter written for The Immunoassay Handbook.  The version in the 4^th edition includes extensive revisions by Jianwen He, based on his experience troubleshooting immunoassays throughout the Asia-Pacific region.

The distressing and puzzling case of missing flight MH370 serves as a reminder that our desire to know all the facts cannot always be fulfilled. Those investigating this incident should be respected and appreciated for the huge amount of work involved, and they must never give up searching for new clues. I am cautiously optimistic that one day they will discover the explanation. Based on my experience, it will not be what we expected.

About David Wild

David Wild is a healthcare industry veteran, with experience in biotechnology, pharmaceuticals, medical devices and immunodiagnostics, which remains his passion. He worked for Amersham, Eastman-Kodak, Johnson & Johnson, and Bristol-Myers Squibb, and consulted for diagnostics and biotechnology companies. He led research and development programs, design and construction of chemical and biotechnology plants, and integration of acquired companies. Director-level positions included Research and Development, Design Engineering, Operations and Strategy, for billion dollar businesses. He retired from full-time work in 2012 to focus on his role as Editor of The Immunoassay Handbook, and advises on product development, manufacturing and marketing.

Interested in more articles from contributors to The Immunoassay Handbook? View those here!

Interested in other articles on Malaysia Airlines Flight 370? View more here.