A person has a COVID-19 test administered at a walk-up...

A person has a COVID-19 test administered at a walk-up testing site on Wednesday in New York City. Credit: Getty Images/David Dee Delgado

Hope is not an effective strategy for dealing with a pandemic.

Two years in, we should all know that. Yet I've been watching with concern bordering on alarm as the omicron variant has entered the country and started transmitting in the community. It has become the dominant strain, taking over from the delta variant (itself no slouch) in a matter of days in some places. If it's not in your neck of the woods yet, it will be soon.

Alongside this there has steady tide of coverage and commentary suggesting that omicron causes mostly mild disease - the implication being that it's not much to worry about, that we need only stay the course and we can ride this one out, too.

But that's premature. Let me be clear: I'm not stating definitively that omicron has some grim future in store for us. I'm saying that there are red flashing warning signs, that we underestimate this virus at our peril and that even the best-case scenario is still bad.

Barely a month after it was discovered, there's still quite a bit we don't know about omicron. The three key areas to focus on are transmissibility, disease severity and immune evasion.

It is clear from data emerging around the world that omicron is highly transmissible and spreads more quickly than delta, which has caused enormous waves of infection in the United States and other nations, including in the parts of southern Africa where omicron was first reported.

Genome sequencing in South Africa, Great Britain, Denmark, Norway and other countries make clear that when omicron takes root, it takes off really fast. We are already starting to see signs of that in the United States: The Centers for Disease Control and Prevention reported Monday evening that omicron made up 73 percent of U.S. coronavirus cases between Dec. 12 and Dec. 18.

If the disease caused by this variant is mostly mild (and don't hang too much on that "if" yet), we could avoid the worst imaginable outcomes. It is important to remember that vaccination, especially among those boosted, protects well against severe illness. Unvaccinated people in South Africa and the United Kingdom have been shown to be at far higher risk of getting hospitalized than vaccinated ones.

But a "mild" case of covid can still make you feel miserable, even bedridden, for days. Huge numbers of "mildly ill" people unable to go to work or school can cause enormous societal disruption, especially while we're already experiencing labor shortages and supply-chain disruptions. There's also the risk of long covid, which can cause physical and cognitive issues for many weeks and months after recovery from the acute phase of illness, and which we still don't understand that well even two years into the pandemic.

And finally, there's math. Let's pull a number out of thin air for demonstration's sake and say that only 2 percent of omicron cases are severe enough to cause hospitalization. Good news, right? Not if omicron also causes exponentially more infections in a condensed time. Two percent of a huge number is a very large number indeed. With hospitals across the U.S. already strained to the breaking point, the implications are frightening.

So how severe is omicron, relative to delta? We haven't seen crushing waves of hospitalizations in the countries where omicron took hold first, such as South Africa. That is good news because it probably means omicron is not causing significantly more severe disease than other variants. But it does not mean we're out of the woods on severity.

Time and again throughout this pandemic, everyone from TV talking heads to politicians have gotten tripped up on the simple and immutable fact that hospitalizations lag infection. It can take weeks for an infection to progress to the point where the patient needs to be hospitalized. Omicron only burst onto the scene four weeks ago. Not enough time has passed for us to feel we have a firm grip on disease severity.

That's because, again, of math: When a variant spreads extremely rapidly, as omicron seems to do, it can send off false signals of reassurance on disease severity.

You've likely seen graphs from several countries showing that the number of omicron cases has skyrocketed in a matter of days. By contrast, hospitalizations have barely ticked up. Perhaps those graphs have been presented side-by-side with charts showing that by the time delta variant caseloads were this high, hospitalizations were already starting to climb dramatically. That may seem reassuring. It is not. Remember: It has still been at most a week or two since almost any region in any part of the world started to see omicron infections in any significant number. You would not expect hospitalizations to follow this quickly. That's not how covid works. The comparison with delta is misleading because it took delta longer to reach a critical mass of cases. We thought delta was fast, but it became dominant over a couple of months. During that prolonged period of case growth, a certain percentage of people who had been infected early got sick enough to require hospitalization. With omicron, we are not talking months but weeks; the case growth has been so compressed, there hasn't been time for disease progression in those who were sickened early on. We need to wait two or three more weeks until we can make even a tentative judgment on disease severity. And by that time, many more people will have been infected.

The science on the ability of omicron to evade immunity has been moving incredibly quickly. Recent data from a large South African study, suggest that the two-dose regimen of the Pfizer vaccine offers 33 percent protection against infection and 70 percent protection against hospitalization. A booster will likely raise those numbers.

That means it seems clear that breakthrough infections in the fully vaccinated will be more common with omicron than they were with delta. That is not good news, for all the reasons I outlined when talking about high transmissibility. It does also appear, as we would expect and hope, that vaccines remain strongly protective against severe illness. But that finding unfortunately carries a couple of caveats.

As I noted above, not enough time has elapsed since omicron first hit the scene, even in South Africa, for us to get a full picture of hospitalization risk. And covid has swept through South Africa several times in the past two years, so a very large proportion of the population has prior immunity from past infection with other variants. Its population is on average younger than in most other countries, so not as likely to develop severe disease. This could be an important factor in blunting disease severity. All that means that South Africa's experience may not be applicable to other populations - notably, the reservoirs of people in the U.S. who remain unvaccinated and who have not yet contracted covid in any form.

So what does all this mean?

It means we must be vigilant. We must take all the precautions that have become so familiar, and yes, so wearying. Get vaccinated. As soon as you're eligible, get boosted. Wear good masks in crowded indoor places. Keep a distance when you can. Wash your hands often.

And on a national level, it is past time for a clear-eyed rethinking of our pandemic policies.

Omicron's very existence reminds us that this virus will keep mutating and will continue to pose grave threats for years to come. It is past time to rally the world to deliver billions more vaccine doses to the developing world. It is past time to make rapid tests readily, freely, widely available. It is past time to demand clear and forthright public health communication from officials at every level. And it is past time to develop proactive strategies for living in a world where covid is continually circulating and we work to minimize its toll.

We have come a long way since the start of the pandemic. Vaccines will protect against severe disease in many people, and we can soon expect to have new therapeutic options in our arsenal to combat the virus in those who do need hospitalization. But the dizzying pace of omicron's advance creates problems all of its own if many doses of those therapeutics are needed more quickly than they can be manufactured.

In March 2020, the threat of the virus was indistinct. This time, the leaps and bounds we have made in testing and genomic surveillance mean we can see it coming. It is just moving quicker than anyone could have imagined - and quicker than any of us can pivot to respond to it.

Hope is deeply important to all of us as human beings. But hope is not a strategy. We need to maintain hope in the face of omicron. It just won't be enough on its own.

William Hanage is an associate professor of epidemiology and co-director of the Center for Communicable Disease Dynamics at the Harvard T.H. Chan School of Public Health.