The near-term objective should be to reduce the acute, exponential growth of the outbreak, in order to reduce suffering and the strain on our health-care system. That will require significant effort, but it can work, as we have seen: Hong Kong and Singapore have achieved linear growth of covid-19 cases, staving off the terrifying exponential upward curve confronting Italy and pushing both the infection rate down and new cases out on the timeline.

The United States needs to take note.

This virus is such a threat because it is both highly infectious and lethal, and not enough people are being tested, despite significant recent effort by the Centers for Disease Control and Prevention. By the time cases are confirmed, significant community transmission has likely already occurred. This is a classic tip-of-the-iceberg phenomenon. It’s also akin to looking at a star; the light we see today was emitted some time ago. But the most useful comparison now is to a fire that threatens to burn out of control. It is one we can still contain, even extinguish — if we act.

The best way to put out the fire is a vaccine, but that is over a year away. In the meantime, we must focus on reducing the height of the outbreak curve. This requires coordination and implementation of non-pharmaceutical interventions. School closures, isolation of the sick, home quarantines of those who have come into contact with the sick, social distancing, telework and large-gathering cancellations must be implemented before the spread of the disease in any community reaches 1 percent. After that, science tells us, these interventions become far less effective.

Simply put, as evidence of human-to-human transmission becomes clear in a community, officials must pull the trigger on aggressive interventions. Time matters. Two weeks of delay can mean the difference between success and failure. Public health experts learned this in 1918 when the Spanish flu killed 50 million to 100 million people around the globe. If we fail to take action, we will watch our health-care system be overwhelmed.

Starting now, public health messaging should be framed in light of this clear objective. Community-based interventions are needed to delay the outbreak peak. On this, the 1918 flu taught us a lot. The difference between the steps taken in Philadelphia, which waited too long back then, and St. Louis, which acted quickly, is staggering. Aggressive interventions put off and ease the peak burden on hospitals and other health-care infrastructure. Ultimately, these measures can also diminish the overall number of cases and health impacts.

Consider the actions taken in Italy. On Feb. 20, Italy reported three instances of infection and no known deaths. On Feb. 21, Italy had 20 cases and its first attributed death. Officials implemented interventions, including school closures, the following day and instituted a cordon sanitaire affecting 50,000 people. That’s aggressive, but it was too late. On Feb. 22, Italy reported 63 cases and a second death. A little more than a week later, there were 2,036 cases, with 140 patients in serious condition and 52 deaths. Today, the numbers continue to climb, with more than 9,100 cases and 460 dead, and on Monday the government expanded travel restrictions to the entire country.

By contrast, Hong Kong and Singapore acted immediately and are still holding the line, literally. Through isolation, quarantines, contact tracing, canceled gatherings and widespread surveillance, they have achieved linear growth of the virus, meaning a reproduction number close to one. What they are doing is working.

Working parents without child care have a legitimate concern, and we must find ways to help one another. But school closings can be the single most effective intervention. Amid an influenza pandemic, schools would be closed to protect the students themselves. Because children are not among the groups most vulnerable to coronavirus, schools should be closed in an effort to reduce community transmission and to protect the children’s parents and grandparents. How long? Epidemiologists suggest eight weeks might be needed to arrest this outbreak. Administrators, students, teachers and parents need to get busy figuring out how to continue the education of our children while contributing to this community-wide public health effort.

The United States and other liberal societies must mount a significant, coordinated response with public buy-in. Panic must, of course, be avoided. Most people who become infected are likely to get what feels like a mild case of seasonal flu. Many will not develop symptoms. But the elderly and otherwise infirm are at risk, and the number of Americans likely to be hospitalized and the subset of those who will require some form of critical care could still be significant.

The rates will be worse if the disease is not aggressively countered early. But I know we can all work together for the greater good.