It is a basic instinct of infectious diseases clinicians and researchers to seek and prescribe prophylactics or treatments that are almost guaranteed to benefit the patients who entrust us with their care. With Covid-19, this instinct can be counterproductive. The new coronavirus travels through populations too quickly and unpredictably for us to wait to tackle it until we have devised nearly flawless solutions. The widespread implementation of imperfect prevention measures, therapies and vaccines may be the fastest way to get a handle on the crisis. Even those communities in the United States that are faring relatively well against the virus right now are still dangerously close to a tipping point: Infection rates and deaths could shoot up again suddenly, as they did in several states this summer. But there is another possible tipping point, too, in the other direction. With more comprehensive use of even moderately effective prevention and treatment strategies, cases of infection and deaths could decrease substantially within weeks. It would be safer then to reopen schools and relax physical distancing restrictions. There are potential benefits to using inexpensive paper-strip tests to detect coronavirus infections, even though those are less accurate than the standard polymerase chain reaction (PCR.) tests, which can return a positive result for tiny amounts of the virus or long after a person has ceased to be infectious. The paper tests have a rapid turnaround time; if deployed widely and frequently, they could be an effective first-detection tool. And here is a simple, imperfect measure that has already saved many lives: the face mask. A mask, especially one made of cloth, is a primitive block against respiratory viruses, and in terms of efficacy it probably pales in comparison with the condom, the gold standard of barriers for preventing infectious diseases. Measuring a mask’s effectiveness at the individual level — particularly in the real world where use is intermittent and imperfect, and where people wear various types of masks — is devilishly challenging. But based on mathematical modeling, my research group found, as it described in a recent preprint (a paper not yet peer-reviewed), that a mask worn by an infected person that filtered only 50 percent of the virus that person exhaled would lower the chances of their transmitting the virus to someone else by 10-60 percent (depending on how much of the virus the infected person carried at the time). When an infected person and another person are both masked, the chance of a transmission decreases by 40-80 percent. According to our model, even when masking does not prevent people from getting infected, it decreases by roughly 10-fold the amount of virus to which they have been exposed, and that, in turn, may limit the likelihood that they will develop a severe form of Covid-19. When these effects are extended to an entire population, the overall impact can be profound, and when deployed along with other measures, they could mean the difference between case numbers that suddenly skyrocket and the suppression of a local outbreak. Since superspreading events seem to be driving the pandemic, even slightly better masking practices among people who cannot avoid situations that favor clusters of outbreaks — like extended time spent in crowded and poorly ventilated settings — could bring outsize benefits. Marginal improvements in the efficacy of masks themselves could also vastly reduce the number of new cases. Substantial investment should go to designing more protective and more comfortable masks, and marketing them with labeling that describes their level of protection as well as how best to use them. Likewise, the widespread deployment of an even partially effective therapy could place the United States in a much safer position. One of the true failures of our response to the pandemic has been the slow development or testing of antivirals, medicines designed to stop a virus from infecting our cells or limit dangerous levels of inflammation. Only a tiny minority of infected people worldwide have enrolled in clinical trials to date, and many of them had already suffered severe symptoms when they signed up. In an important study of remdesivir, which cripples enzymes that viruses need to replicate themselves, the drug shortened the duration of Covid-19 symptoms in hospitalized patients by about four days. Another study of the effects of dexamethasone, a common steroid, showed a slight decrease in mortality. But with either drug, the studies suggested, only one in every 20 people who received treatment would be saved from dying. This is not surprising. A serious case of Covid-19 is akin to an uncontrolled forest fire. Much of the lung and vascular tissue is inflamed; damage occurs in multiple organs. Treatment is more likely to succeed when it is started in the early stages of infection, when the fire is small and localized. This is the case with viruses such as influenza, Ebola, zoster and HIV. Initiating treatment when the first symptoms of Covid-19 appear would not only help prevent deaths; it could also lower the hospitalization rate, relieving some of the burden on emergency departments and intensive care units. Multiple trials are now underway to investigate early treatment for Covid-19 with antivirals — or repurposed drugs commonly used for other diseases, or antibodies to this coronavirus that have been engineered and mass produced. Yet it’s unclear whether these studies can be completed quickly enough to meaningfully lower rates of hospitalization or deaths before a vaccine is developed and widely distributed. This is partly the case because trial drugs are usually evaluated based on whether they lower cases of hospitalization. But hospitalization rates may not be the only, nor the best, endpoint for clinical trials — not when time is pressing. Hospitalization rates for Covid-19 patients involved in early-treatment trials in the United States have tended to be below 5 percent, which means that a study that hopes to demonstrate a statistically significant difference between a drug and a placebo requires the participation of more than 1,000 people. Endpoints other than lower hospitalization rates could be set for assessing trials; one of them could be determining whether certain drugs shorten the duration of patients’ symptoms. Such criteria would allow for still rigorous but much speedier testing involving, say, fewer than 100 participants. Smaller, nimbler studies would also promote comparison among a greater breadth of promising medicines, all with an eye toward getting effective drugs to the market as quickly as possible. The most crucial area where the search for perfection could come at the expense of the greater good is the development, assessment and licensing of vaccines. As with antiviral therapies, a vaccine should not be distributed to the public without its safety and efficacy having first been demonstrated in randomized double-blind, placebo-controlled clinical trials. But at issue, again, is how we choose to define efficacy. The US Food and Drug Administration typically approves vaccines that are at least 50 percent effective at preventing a disease. But even a vaccine less effective than that could substantially lower the number of cases of coronavirus infection and Covid-19-related deaths, if it were rolled out fast enough and given first to the people most likely to get infected or to infect other people. As others have argued, vaccines don’t just prevent a disease; they can stop the pathogen that causes it from being transmitted. Mass immunization programs benefit not only the people who are vaccinated, but also everyone else, since they are less likely to come into contact with an infected person. For example, the widespread inoculation of children in the United States with a vaccine for the pneumococcus bacterium, a common cause of pneumonia, has been shown to curb deaths and hospitalizations from the disease among adults. Similarly, some of the people most at risk of developing severe cases of Covid-19 — the elderly, the immunocompromised — may not adequately respond to a vaccine. But they could nevertheless be shielded by one if a sufficient proportion of the total population were inoculated with it. And even a vaccine that does not protect against Covid-19 might be of enormous utility, if it causes recipients to carry less of the coronavirus and therefore, presumably, be less contagious. The potentially huge benefits of rolling out low-sensitivity tests that can be administered quickly and frequently is increasingly being recognized as a viable path to, say, reopening college campuses and professional sports leagues. If over the next six months we can also make small, iterative gains with masking and modestly effective therapies and vaccines, then the worst of the pandemic might soon be behind us. For that to happen, though, we must avoid the temptation of seeking perfect solutions. Dr. Joshua T. Schiffer is an associate professor in the Vaccine and Infectious Disease Division at the Fred Hutchinson Cancer Research Center and the Division of Allergy and Infectious Diseases at the University of Washington in Seattle. The New York Times