Why Is Implementation Science So Damn Slow?

I am on vacation this week, so mostly in brain-rewiring relaxation mode, averaging a half dozen naps per day. But when awake, and not distracted by the ocean and kids, I have been noodling a lot on time (while mostly in an horizontal posture).

I am increasingly convinced that one of the biggest problems in academic global health [and implementation science more generally] is not talent or technology, but time.* We routinely spend ?3-5 years generating evidence for decisions that programs make in 5-10 weeks. The entire NIH incentive structure seems to rewards slow, methodical knowledge generation. Which is frankly bonkers.

In a world of real time analytics and shrinking budgets, a lot of HIV implementation science (at least US-funded, I cant speak to elsewhere) is in danger of rendering itself obsolete if it continues to move so slowly. With PEPFAR programs rapidly transitioning to partner governments, HIV programs need evidence now now, not on 3 years time. Which raises a few inter-related questions; how do funders [Gates, NIH] incentivize speed while not compromising on quality? And how do researchers stop optimizing for publications [which take >three years to write], and start optimizing for programs, that need decisions three weeks away?

Eric Goosby and I argue in the penultimate chapter of our book that implementation science has to become dramatically faster. If private sector can move at the speed of programs, then why cant academics?!