We're in for a grim April and a harsh May before we approach some new kind of normalcy.

That's the message that has been imbedded in statistical models, white papers and pronouncements made by governors, mayors and White House advisers over the eerily quiet and brutally long two weeks that stretched from March 16, when Santa Clara County and five other Bay Area counties first ordered residents to stay at home due to the coronavirus outbreak, to this past Tuesday, when the counties extended the order to May 3.

But while the initial Bay Area shutdown seemed jarring, the March 31 extension felt almost inevitable. Between the two orders, the number of confirmed cases in the county jumped from 138 to 890 and the number of deaths from COVID-19 went from three to 30. Bay Area schools officially shuttered for the rest of the academic year; and U.S. President Donald Trump abruptly pivoted from a bullish plan to reopen the country by Easter to declaring on Tuesday, "This is going to be three weeks like we've never seen before."

While health experts are reluctant to answer the elusive question "How long will the outbreak last?" with any degree of certainty or specificity, most measures indicate that Santa Clara County, like the state and the nation, will see a sharp increase in cases in April and early May before things begin to level off.

Health officials have been reluctant to predict the duration of the outbreak, partly because the lack of widespread testing makes it difficult to know how many people in the county are infected. County Executive Jeffrey Smith said on March 24 that he believes that based on modeling done thus far, the number of infected persons is probably at least 5,000. (The county's confirmed number of cases on April 2 was only 1,019, however.)

Testing remains far from widespread, despite efforts to boost supplies of tests by academic institutions like Stanford University and University of California, San Francisco and commercial firms such as Abbott, Roche and Quest Diagnostics.

Some experts estimate the peak in cases could happen earlier or later — in part varying by location. Santa Clara County health leaders say a local surge will start by mid-April, but Dr. Mark Ghaly, the state's secretary of health and human services, projected a peak in the state as a whole in mid-May. Both state and county leaders have refused to publicly release details regarding their modeling, saying they prefer to reserve those models for internal decision making.

The Institute for Health Metrics and Evaluation, which is part of University of Washington, created a model for every state in the nation. On April 3, it predicted that in California, the COVID-19 curve showing numbers of cases will peak on April 26, when the state would see 119 deaths that day related to the virus. Under this model, the number of daily deaths would then gradually go down to 16 on June 1 and then dip to single digits by mid-June.

At best, experts can only take an educated guess based on "assumptions" and prepare for the worst, they said.

"Nobody knows," Dr. Yvonne Maldonado, medical director of infection control at Lucile Packard Children's Hospital and Stanford professor of pediatric infectious diseases and health research and policy, said this week when asked about the projected peak and when the crisis might end.

"Two weeks ago, experts said we would be like Italy. Two weeks have passed and we are not like Italy. On the other hand, we could be like New York or New Jersey," she said.

The virus might be distributed differently across the Bay Area, making predictions difficult, Maldonado said. Some communities may have a higher infection rate, others a lower one. Demographics, levels of health within communities and populations, access to medical care, the movement of people and daily behaviors can all affect how the virus spreads.

There are some officials who are willing to openly predict numbers. The city of San Jose's manager's office reported grim estimates for Santa Clara County at the March 24 City Council meeting, estimating 2,000 to 16,000 deaths and 9,000 to 19,000 people who could be ill in the next 12 weeks out of a population of 1.94 million. County public health leaders, however, distanced themselves in a single-paragraph statement the next day, saying they hadn't vetted the information and were studying it.

On April 1, San Jose Mayor Sam Liccardo voiced support for the data, noting the city manager's office had presented it to Smith prior to releasing it. Emails show the county did not dispute the data so long as it was clear the numbers were based on assumptions. And those estimates and assumptions were fundamentally correct, Liccardo said, even as City Manager David Sykes conceded the county and the city were not completely "in sync."

Despite those worrisome projections, county health leaders say they are starting to see signs of hope. Local hospitals are not yet nearing their capacity and are not seeing the types of shortages of personal protective equipment needed by health care workers as in New York and Michigan.

Dr. Dean Winslow, professor of medicine in the division of hospital medicine, division of infectious diseases and geographic medicine at Stanford University School of Medicine, said Tuesday that about 33 patients at Stanford had been admitted for suspected COVID-19 infections. About half are confirmed cases; a smaller number tested negative. The rest of the results are still pending.

And among the county's 11 hospitals, which have 1,475 beds, there were 152 confirmed COVID-19 patients and another 90 who were suspected of having COVID-19 on March 30, according to Smith.

On March 31, even as county Health Officer Sara Cody announced the extension of the stay-at-home order and added new restrictions — including a clamp down on nonessential construction activities and new requirements for businesses to post plans for keeping their facilities safe — she said there are signs that the order is working and that the spread of the virus is slowing.

"We have some signs, some very soft signs, but I think the train is beginning to slow down a little bit," Cody said.

"What we need to see is that our demand curve — which is how many people are ill, requiring hospitalization and ICU care — comes to a place where it's comfortably nestled under our supply curve," Cody said, referring to the supply of beds, staff and medical equipment that is needed to properly take care of the patients.

"It's a complex balance. I want to say that bringing in the data to understand all the needs in our health care system — be they supplies of beds, or staff or personal protective equipment — is fairly complex, as is understanding the trajectory of the epidemic in our county."

Looking at the coming months

Any estimate of when the pandemic will end is inherently fuzzy, given that the determination depends on myriad variables, including availability of testing for COVID-19, hospital capacity, and effectiveness of social-distancing measures.

Andy Slavitt, who served as acting administrator for Centers of Medicare and Medicaid Services during the Obama administration and who now heads the organization United States of Care, suggested in a March 28 post on Medium that social-distancing measures would probably need to remain in place through May or even into June to slow the spread of the virus.

"If we buckle down (6 to 10 weeks?) with social isolation, the curve flattens and, in a really strong effort, can decline. If we let up, we are in for a very rocky and lethal extended period of time," Slavitt wrote.

A recent analysis by National Geographic of how different cities handled the 1918 influenza pandemic shows the devastating impacts of halting social-distancing measures too soon. San Francisco did so immediately after its count of daily casualties dropped, only to see a second wave about a month later. It ended up with 673 deaths per 100,000 residents, according to the analysis. New York City, which kept its measures in place for longer, had 452 deaths per 100,000 over the same period.

"A delay or letting up a little early on social distancing means lives lost," Slavitt wrote, pointing to the data.

Stanford health experts echo that sentiment.

"We really need to be very careful not to relax measures too early," Winslow said.

Testing has lagged throughout the state, and that's a major concern, he said. Asymptomatic people who are still carriers are being neither tested nor tracked. There's also no significant understanding of who is potentially already immune — two pieces of critical missing data to understanding the outbreak, he said.

Winslow and Maldonado said predictive models to determine if there could be a resurgence of the disease also aren't reliable at this point, and they urged caution.

Some analysts see a way we can ease out of the current economic and social crisis while also keeping an eye on public health.

A new white paper from the American Enterprise Institute, a conservative-leaning think tank, proposes a phased "roadmap" for getting out of the pandemic. In the first stage, governments impose social distancing measures such as bans on public gatherings, stay-at-home advisories and isolation of COVID-19 cases either at home or at hospitals, allowing communities to flatten the curve. During this time, the health care system ramps up its capacity and obtains life-saving ventilators, personal protective equipment and other critical medical tools.

The paper's authors, a list that includes three former FDA officials as well as Johns Hopkins Center for Health Security epidemiologist Caitlin Rivers and Johns Hopkins health security expert Crystal Watson, propose remaining in this stage until there's a decrease in cases for 14 consecutive days; hospitals are safely able to treat all patients that require hospitalization; and each state has enough tests for people with COVID-19 symptoms. The paper estimates that the country would need to produce about 750,000 tests per week — a number based on the testing ratio in South Korea, a nation lauded for its quick and effective response to the pandemic.

Once these goals are met, the nation can move to the next stage, in which schools can reopen and most people can start going back to work. Physical-distancing restrictions would be gradually lifted, though not eliminated, until the third stage, when a vaccine is developed.

(At a Wednesday news briefing, Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Disease and top White House adviser on the pandemic, called the development of a vaccine "the ultimate solution" to keeping the coronavirus from coming back.)

The American Enterprise Institute paper argues for a slow and gradual progression between phases, done on a state-by-state basis. Furthermore, a state that reopens should revert to the first phase "if a substantial number of cases cannot be traced back to known cases, if there is a sustained rise in new cases for five days, or if hospitals in the state are no longer able to safely treat all patients requiring hospitalization."

In other words, states may need to go through another cycle of social-distancing measures if the nation has not yet developed and scaled up production of a COVID-19 vaccine.

Getting to the end point

Stanford University biologist Erin Mordecai and her team have likewise been modeling scenarios for the critical period between June, when the surge is expected to abate, and July 2021, when a vaccine would presumably be in place.

But like the American Enterprise Institute, Mordecai's team sought to identify ways to get society back to its normal rhythms before the vaccine is created.

"The idea of a shelter-in-place for a year or more doesn't seem feasible," Mordecai said. "We were interested in a possible alternative to just having everyone stay home for a very long time."

Their modeling suggests that the most effective approach may require repeated social-distancing interventions, with strategies quickly adapting to the situation on the ground.

The models also indicate that launching social-distancing interventions early in the epidemic cycle is more critical than the severity of the social measures imposed.

Hence she and her team created the "lightswitch" approach. Under this concept, social-distancing measures are gradually added or removed over the course of the year, based on the fluctuations in numbers of COVID-19 cases. On one end of the scale are strict interventions like quarantines and fines for people who flout stay-at-home orders. On the other, public life starts to open up while restrictions remain in place for vulnerable populations and people with COVID-19.

"You can turn on and off social-distancing interventions over time," Mordecai said.

Such an approach, she said, would require "very careful surveillance," enabling authorities to quickly intervene if the number of hospitalized COVID-19 patients creeps up.

"We can use that as a benchmark. Do you exceed hospital capacity? Then we may need a shelter-in-place. Will hospitalizations go down? Then we can resume normal activities."

(The team's interactive website, covid-measures.github.io, allows users to test out intervention strategies by plotting two — one after the outbreak and another one later in the year — and seeing the impacts of these strategies on hospitalizations, infections, recoveries and deaths.)

While much is being discussed about effective strategies for handling the outbreak and the best ways to get us at an endpoint, health experts can only speculate about whether the new coronavirus will mutate and continue to plague our lives.

Maldonado noted that COVID-19 is related to six other coronaviruses, including the common cold. Some of those mutate over time and come back; others don't.

The 2003 Severe Acute Respiratory Syndrome (SARS) coronavirus, which sickened 8,000 people and killed nearly 800, never came back after 16 years, although some thought it would. The 2012 Middle East respiratory syndrome-related coronavirus (MERS) did reappear, although not at a high level because it was transmitted only through contact with camels.

COVID-19 is highly contagious, which gives Maldonado pause.

"It's highly possible we'll see it again," she said. "That's why we need to develop antivirals and a vaccine."

Find comprehensive coverage on the Midpeninsula's response to the new coronavirus by Palo Alto Online, the Mountain View Voice and the Almanac here.