The stumbling block facing “bench-to-bedside” translational research is often represented as the difficulty of implementing consistently across whole neighborhoods a program that has been shown to improve children’s lives in trials.

Some branches of medicine have it easier: three million doses of influenza vaccine will work as well as 300 doses and cumulatively may be much more effective.

But other medical interventions must deal with familiar uncertainties and complexities. And, given an injection of lateral thinking, some experimental solutions, which resist the conventional distinction between the bedside and the bench, may be more universally relevant.

Fidelity, dosage, adaptation and the need for a “manual” all figure in different guises in the New York Times magazine’s recent account of “delivery research” activity in Salt Lake City.

It centers on the work of former Harvard cancer specialist Brent James, now executive director of Intermountain Healthcare and architect of a training program that combines statistical methods, management theory and, arguably, elements of open design.

One case study focuses on the development during the late 1980s of ventilator treatment for acute respiratory distress syndrome (ARDS) – a killer nowadays associated with the complications of swine flu. It sounds a far cry from any parenting program, but there are illuminating parallels in the narrative.

The pulmonologist concerned, Alan Morris, was worried that his trial might be undermined by a tendency among doctors to adjust ventilators differently for similar patients. Intuitive “twiddling of the knobs” was something he did himself. It was an aspect of bedside expertise.

So he and Brent James decided to write a treatment protocol. “Some of the recommendations were based on solid evidence,” the NYT reports. “Many were educated guesses. The final document ran to 50 pages and was left at the patients’ bedsides in loose-leaf binders.”

Morris’s professional colleagues were skeptical, but their resistance was lowered by an invitation to depart from the “manual” whenever they wanted. The point of it was to reduce variation, not to eradicate it, and to enable the those monitoring the trial to isolate the aspects of treatment that made a difference.

“While the pulmonologists were working off of the protocol, Intermountain’s computerized records system was tracking patient outcomes. A pulmonology team met each week to talk about the outcomes and to rewrite the protocol when it seemed to be wrong.

“In the first few months, the team made dozens of changes. Just as the pulmonologists predicted, the initial protocol was deeply flawed. But it seemed to be successful anyway. One widely circulated national study overseen by doctors at Massachusetts General Hospital found an ARDS survival rate of about 10 percent. For those in Intermountain’s study, the rate was 40 percent.”

The results have been contested and debated and the NYT says Morris himself has been reluctant to give the protocol credit for the improvement.

But, from Brent James’s point of view, the fact that medical staff working collaboratively in the field managed to put together a complex set of useful clinical guidelines was a significant achievement.

Other similar exercises followed, relating to 50 clinical conditions said to account for more than half of Intermountain’s patients. In each case a committee of doctors, nurses and administrators tried to identify variation and to figure out beneficial refinements.

The guidelines are embedded in the hospital’s computer system alongside patients' electronic records. By combining the two it has been possible to track and analyze patient outcomes.

“Doctors with consistently poor results can expect to be pulled aside for a collegial conversation with a supervisor about what they might be doing wrong. Doctors with the best results can expect to be asked what they are doing right,” according to the NYT report.

Everyone involved appears to concede that the process is rarely easy or clear-cut. Nevertheless, tracking outcomes and adjusting care proceed transparently in parallel.

By an optimistic reckoning, bench and bed begin to look as if they might fit together side by side in the same exploratory space.