Efficiency: Paradoxical Impacts of Technology
Southwest Airlines gone awry
We dedicated this portion of our professional lives to improving healthcare delivery. In every other area of our lives, we enjoy faster and more convenient service, better quality products and services, all enabled by supply chain management, attention to logistics, and the availability of Web services that globally connect our browsers, iPhones, BlackBerries and Pre(s).
Last week, I donned a backpack, walked a mile to the grocery store, picked up a half dozen items, bar coded them as I shopped and put them in my backpack. When I arrived at checkout, the items stayed in my backpack. I bar coded the checkout scanner and my complete receipt of items appeared, along with all of the “frequent-shopper” discounts. I swiped my credit card, signed the pad and resumed my walk home. The entire checkout process was less than 30 seconds. No lines. No conveyor belt. A checkout clerk nearby offered to assist, but it wasn't necessary.
Bar-code self-check-out ( Too many Atkins' low carb bars?)
Of course, the milkman/breadman of my youth has now been replaced by the webvan. We place our main weekly grocery list online and it's delivered. No need for the walk, unless I need the exercise ... which I do.
The relevance to healthcare is obvious to everyone reading this. We've got bar code medication administration, which coordinates the final step in delivering medication to inpatients, delivering easier documentation, the five rights checks, documentation as a by-product of process, and other kinds of clinical decision support (e.g. a final opportunity to hold a heart slowing drug to a patient whose heart is already too slow).
Today, I'm flying from Baltimore's BWI airport on Southwest Airlines. Southwest, we all remember, drove its costs down more than any other airline in the last 10 years, by taking all of the waste out of its processes. It's a great story and many of us have enjoyed the good parts.
When I went to check-in online, however, this time I received the message that online check-in was filled up and I should check in at the airport. I just wasted the 10 minutes it took me to get to that point.
When I got to the airport, the kiosk gave me a security clearance card but not a check-in. It advised me to check in at the gate.
With each of these touch points, Southwest had the opportunity to eliminate a downstream bottleneck for me and for them. And yet, their software actually set me up to stand in another line. A "no value add" line.
The experience was clearly inferior to the other airlines. I've never been denied an online check-in before. I've never had a kiosk refuse to assign me a boarding pass.
All of the "happy talk" on Southwest's Web site about fast, friendly service sent me the opposite message than they intended.
As we race to improve the efficiency and effectiveness of care coordination and delivery, we need to remember that automation can, paradoxically, make things slower and more expensive. We need to carefully design, analyze, test, and monitor our new processes. These aren't simply theoretical or rare occurrences. It's well known to network engineers that adding capacity to a network can make it slower. One such effect is called Braess's paradox and is elaborated here: http://en.wikipedia.org/wiki/Braess%27s_paradox )
Braess's paradox, credited to the mathematician Dietrich Braess,
states that adding extra capacity to a network,
when the moving entities selfishly choose their route,
can in some cases reduce overall performance.
Another example comes from India. A colleague who recently returned said that after they added a new central overpass bridge, the traffic congestion clearly got immediately worse.
At least the technology now allows me to write these blog posts while I'm waiting in line, from anywhere!
A footnote (from Wikipedia article above): How rare is the Braess Paradox? It's about a 50/50 proposition.
For example, in 1990, the closing of 42nd street in New York City reduced the amount of congestion in the area.