Paul Dvorak

It seems the FDA just can't catch a break. Almost on the heels of releasing the improved MDUFMA agreements, an updated list of fees the agency charges OEMs to review new products before allowing commercial introduction, the agency has come under harsh criticism from within. A subcommittee of the FDA's Science Board reports that regulatory systems overseeing U.S. drugs and devices are at risk, along with the U.S. food supply. The subcommittee attributes the deficiencies to soaring demands on the FDA and resources that have not risen in proportion to those demands.

Although the committee should be applauded for its work, its recommendations from intelligent people (three of them former FDA commissioners) fall short of real change. It suggests more money, more staff, and updated IT. Any business guru would remind subcommittee members that doing things the same as before will not produce different results. In fact, only one of the three recommendations, modernize the IT, sounds likely to have an effect.

But better yet, rather than modernize the IT, why not outsource the IT to a private company that wins the contract in closed bidding? Like the rest of science, IT is progressing at a rate that demands attention from specialists who can keep up. So why not let the FDA focus on approving devices and drugs, rather than wondering whether the email is working, a frequent concern according to the Board's report. It's bad enough that the report cites a loss of FDA data from recent email failures. It's worse that “Critical data reside in large warehouses…in piles of paper documents. There is no backup for these records which include clinical trial data.” That means the FDA cannot even do computer searches on data it demands companies submit for product approvals.

A rational person living in year 2008 might wonder why it is necessary to ship paper reports to the agency? A more logical idea is to keep the data on OEM servers with access only by FDA staff. After all, Google Inc. solved the problem of searching large databases long ago.

The other two proposed solutions need even more radical rework. Adding people and upping the budget can't be serious proposals for the agency's burgeoning problems. The 2008 budget, for instance, hits $2.1 billion, a 5.3% increase over 2007. A budget breakdown (fda.gov/oc/oms/ofm/budget/2008/summary.html) makes no mention of IT spending, although the agency is moving to a new multimillion dollar campus which will bring together workers from 20 different sites. In a few years, the agency will be right back where it is today. Why? Because the world's bounty of knowledge is nearly doubling annually and will quickly outpace any additional capacity.

Perhaps a more obvious idea would have the FDA work with its counterparts around the globe and divide up the work load. It would certainly make sense to outsource some tasks. The approval of medical devices, for example, could be handed over to private companies like UL.

And for tasks kept in-house, adopt lean manufacturing principles to speed up the work. In the same way lean principles eliminate waste on shop floors, they can eliminate waste in office work. Lean efforts would be most difficult to implement because they would be radical departures from the norm. Failure is an option because a lean reformation calls for buy-in and commitments from supervisors and employees. That means doing things differently.

Inefficiencies arise in government agencies because there is little impetus to change. Yes, a lean reformation is easier said than done, but I'd bet there are “radicals” inside the FDA just itching to rework procedures.