THE CURRENT approach to the coronavirus pandemic in the United States is based on wishful thinking — that a vaccine or drug therapy will be available by the end of the year, or sooner; that death and illness will taper off with the summer heat, and not come back next fall. But what if none of this happens? What if the novel coronavirus sticks around for a year or two or longer? In that case, diagnostic testing will be critical to our ability to manage lives, jobs, schools and health. Yet we still lack a federal strategy to get there.

Diagnostic testing is important, absent a vaccine or therapy, as part of a concerted effort to identify the sick, isolate and treat them, and allow everyone else to get back to business. Right now, testing is the foundation of state decisions about reopening, yet the testing landscape is disorderly and inadequate. After a miserable start, the pace of testing is slowly ramping up, now exceeding 400,000 daily. But that is still far, far below what experts say would be required to sustain a new normal. The effort has been left to 50 states and a hodgepodge of academic laboratories, hospitals and private companies. Some laboratories are overwhelmed, and others underutilized.

Last week, the Atlantic reported that some states were mixing reports of diagnostic testing — finding those who are currently infected — with results from serological testing, a different test that checks for a past history of infection. The mistake affected results in Pennsylvania, Texas, Georgia and Vermont. Virginia was also mixing them, but reversed course when it was discovered.

Another worrisome question mark is testing accuracy, underscored by the questions raised about Abbott’s ID Now machine being used in the White House and elsewhere around the country. The equipment, about the size of a toaster, is the kind of quick point-of-care test that could make diagnostic testing more commonplace, but last week concerns surfaced about whether the machine is giving false negatives — wrongly indicating that persons are not infected when they may be.

Researchers believe testing for covid-19 antibodies in plasma can lead to possible treatments for the virus and help indicate when to reopen society. (John Farrell, Jonathan Baran/The Washington Post)

The Center for Infectious Disease Research and Policy at the University of Minnesota pointed out in a new report that a testing system must succeed at lining up a cascade of essentials, all of which have been scarce lately: chemical reagents, testing equipment and supplies, and skilled operators. Already this is proving difficult. The report called for prioritizing those such as first responders and health-care workers — as well as the elderly, who need tests the most — and “a smart approach to testing that requires the right infrastructure, right population to test, the right test, and the right application of test results.” So far, that goal is elusive. And, on top of that, we need to test ever more widely, and we need far greater capacity.

President Trump eschewed an overarching federal response to the testing challenge. But it is not too late to consider a Manhattan Project-size national effort on diagnostic testing. We have the technology, understand the problems and should spare no cost. The ideal outcome would be to find, in January, that we have a vaccine and have “wasted” some money on testing infrastructure. But if we fall short of that ideal, and face the prospect of losing trillions more in economic output, the billions spent on testing will turn out to have been a cheap and essential investment.

Read more: