Tuesday, April 17, 2012

How History Taking and a Toyota Mechanic beat a Diagnostic Computer

If you give enough time to an experienced clinician working in an ambulatory setting, what percentage of acute complaints would s/he be able to diagnose correctly with just taking a good history?  What about with history plus physical examination?

The answer is about 50% with history alone and about 80% with history and exam.  Yes, s/he would probably order some confirmatory tests or try some medications as empiric treatment but the above numbers would hold up.

The importance of history taken by an experienced diagnostician was borne out recently.  A warning to regular readers of my blog, this is another car story.  Don't tell me I did not warn you!

I have a 1999 RX300 which was bought mainly to combat the snowy NE Ohio winters.  It has been a reliable vehicle and I fully intend to keep on driving it, till it falls apart.  You will probably accuse me of being an emotional fuddy-duddy but it holds a special sentimental value for the family.  So it was particularly upsetting when last year, early in winter it started to give all kinds of problems.

I took it in to the place where I get the oil changed and they are really nice folks, polite and accommodating.  The check engine light was on, and they queried the car computer, read the code, looked up the computer and told me that some sensors needed to be changed.  We did that but within a day the engine light came on again.  This time the mechanic told me that the transmission was gone and it would cost more than the resale value of the car to fix it.  I was crushed but appreciated the fact that he did not make me spend a ton of money before telling me this.  I began to look for someone who would buy it.

Then a friend of mine recommended that I speak to this guy who works at a Toyota dealership.  The Toyota Highlander is almost exactly the same vehicle as the RX300.  So he might be able to tell me more.  So I gave him a call.  I fully expected him to read him the codes from the computer readout.  Imagine my surprise when he asked me to describe what the car was doing!

I told him how I had gone abroad (a workshop I did for physicians in Singapore in October) for about 2 weeks.  Right after I came back the car started misbehaving.  It would work fine for the first 10 minutes or so and then when I tried to accelerate beyond 40 mph it would start revving up like it was stuck in a lower gear.  I could not go on a freeway for fear of this.  He started laughing and asked me if we had seen any rodents in the garage.  I felt like I was talking to Sherlock Holmes!  My wife had told me that she had seen a rodent near where we kept the dog food bags.

So he explained.  The Highlander and the RX300 have an engine intake area that rodents love to nest in.  If the car is not used for a while they start nesting there.  This is particularly true of the fall season as they prepare to hibernate.  The intake area is close to the wires that run to the knock sensors.  The rodents eat the rubber on the wires and this shorts out the sensors (or something like that).  The guy to whom I took the car to first, read the computer code for the knock sensors being faulty and changed them without realizing that the problem was caused by the wires.  Thus he replaced the sensors but the did not fix the cause.

Long story short, (well not really but it was a pretty cool story) the Toyota mechanic changed the wires and the sensors and the car now drives like new.  The key portion of history was that I did not use the car for a while during the nesting season, that we had rodents, that the problem was same as that caused when a knock sensor is faulty.  His experience with having seen this before due to working on similar cars in NE Ohio for years helped him recognize the problem.

This is a story I will tell all my trainees.  How a well directed history taken by an experienced clinician can beat multiple tests and technology!