Why Instructions Matter…

In 1970 my wife, then 12, was stung by a bee.  The ensuing reaction caused her leg to swell such that she could not walk for 6 weeks.  Since that time she has been understandably very careful around bees as the next time she got stung it could (literally) mean death.

Yesterday, I left about 5pm to get the dog from day care when my wife called.  She was at home by herself watering plants in the back yard, when she'd been stung by a bee on her big toe.  She was, needless to say, quite panicked. She has an epipen, but until this critical moment she had never had to use it.  We have the preconditions: high anxiety, very time sensitive, very low familiarity with device, but extremely important.  She pulled out the patient instructions titled “DIRECTIONS FOR USE” on one side and “PATIENT INSERT” with the heading “IMPORTANT INFORMATION” on the other.  It was multiple colored with greens, orange, and black.  There was no obvious starting point.  There are three illustrations, but they are quite small and not instructive.  I could hear her as she started to read them out loud.

The instructions on the patient insert had

• Ten(!) steps
• Three illustrations – None instructive
• Confusing redundancy – Was there a reason for the repetition?
• Font that was too small – My wife wears reading glasses which could not be found in the heat of the moment, and
• Unnecessary complexity for the emergent state of usage - Health literacy is a major concern.  According to the Institute of Medicine 90 Million people in the US have difficulty reading and understanding even simple health information on prescription drugs.  I ran a Gunning Fog index of readability which is an indication of the number of years of formal education one requires in order to easily understand the text on the first reading.  The instructions were at just under 9 years of formal education.  Furthermore, the instructions do a very poor job of emphasis.  Which is more important: font size, capitalization, italics, bold, underlining, color, bold + caps, centering, etc?  It is very hard to tell with this document as there is no principled use of emphasis – no sense of priority.  Again, some of these things simply don’t matter much in normal day to day, but in a panic state it matters a lot.

On the pen itself there were instructions (in 3 steps) that were quite simple.  Again, the font was too small.  But there’s no way to grab the pen and read those instructions at the same time.  And even the idea of looking on the pen for instructions (where you actually grasp it) was not immediately apparent.

In short, it would be very easy for a person in her state to incorrectly inject such that life was at risk.  On top of it, many would be squeamish about doing a self-injection no matter what state they're in.  In the end, she got the dose but the high anxiety of the sting coupled with the poor instructions accompanying the device had immediate and serious implications. 

Most epipen holders are ‘trained’ by a nurse in the office when they are prescribed the device.  Most people will politely listen and ask questions.  Few people get to practice with the device at that time.  I would bet that virtually none will practice under conditions of fear when the medication is truly needed.

I don’t know how many people fail to operate their epipens correctly or die because they can’t self inject, but I’m guessing it’s greater than zero.  I’m not challenging the efficacy of the medicine or the device – although there could be improvements to the device.  I am challenging the simplest – yet most often overlooked part – the instructions.  The instructions are the link between the device and the user.  These are not complicated devices, but imagine that you could become incapacitated in minutes and perhaps even die if you don’t get this medication.  If the instructions fail to properly direct even 1% of the user group – is that acceptable? 

I’m sure the company that makes the pen spent thousands of hours and hundreds of thousands if not millions in development and engineering then testing and retesting.  I’d bet that they spent less than 1% in testing the instructions with real people under any conditions (let alone high stress).

In the end, she got the medication with the help of a neighbor who’d had experience with pens.  The shame of all this is that this situation is not unique.