the gap between stimulus and response

quotes from the hospital, and humans as the instrument of data capture

We were at Frjtz in Hayes Valley on a Sunday morning, when Shan told me I had a “type.”

“What do you mean, I have a ‘type’?” I was indignant.

“What do you mean you don’t have a type? You totally have a type. Think about it. Tall. Skinny. Athletic. Nice. Weird.”

This guy is not my type. Creepy airport randars do not fall in those parameters.

Ouch. I couldn’t even think of an example to refute with. If you ask me, I’m completely diverse in the people I choose to hang out with. My friendships and relationships span the gamut of cultures, upbringings, ideas, and backgrounds. I like people who are intelligent, nice, funny, interesting, adaptable. Right?

The truth is, humans are terrible at articulating what it is they like or want. So goes the old adage: actions speak louder than words. Whether it’s what you do in your spare time, your dating history, or how you use a medical device, it’s universally true. And if you don’t believe me, sit down for brunch with your best friends and ask them what they think of your past relationships.

On Thursday, at Phoebe Putney, I asked a nurse which modality of measuring body temperature in the ICU is most accurate. “Core temperature - so either rectal or bladder.” (PP doesn’t use esophageal probes in ICU or floor units - only the OR)

“OK - so if you’re running a hypothermia protocol, and you’re using two modalities - oral & bladder-based - you trust the bladder based first?”

“Yes.”

Five minutes later: “I had this patient last week that went septic. I had a temp Foley in him, but his temperature seemed really low. It shouldn’t have been that low. I put in an oral thermometer, and it gave me a normal reading. I figured, that Foley must have just been lagging.”

Wait a second. She just told me, five minutes prior, that the Foley reading - the bladder-based temperature - is the most accurate reflection of what’s going on inside a patient. But then, when it came to taking care of a patient, she didn’t like the number the Foley gave and put in an oral thermometer - like you get at CVS - and took the measure.

When we do design research for product development, asking questions is a good start. But it’s only a start, and doubly so if you’re developing something new. The only way to drive behavior, is to observe and understand it. That means: don’t rely on focus groups, or interviews, to give you the information you need to design a successful product.

Enter design ethnography.

Ethnography, if done right, is exhausting. Wendy Newstetter put it best: “You should be drained after four hours, because you are the instrument of data capture.” It requires a basic understanding of the environment you’re in, but a perpetual sense of “newness” or naïveté. You have to dwell in conscious incompetence: you know how much you don’t know.

But it’s worth it. I spent some time on the phone with Jane Chen of Embrace last week, and she told me the biggest learning so far after launch has been that you can never know your users enough. They will “continue to surprise you in the way they approach the problem and the product,” even after - especially after - it’s been put in their hands.

Alas, although I’d really like to videotape our simulated use sessions, HIPAA dictates that this is the closest I can get to taking pictures inside the hospital.

We took our (rough) prototypes to Phoebe. We got better feedback in the first 60 seconds of letting the nurses simulate use, than we did in the 15 hours of clinical IDIs we’ve done this month. It doesn’t matter that they weren’t polished - you can’t be afraid of letting someone in the sandbox before you’ve finished your castle. After all, they’re the ones that will be living in it one day.