The spread of monkeypox is slowing in the US.

(CNN) — The pace of new reported monkeypox cases in some major cities, and in the United States in general, has started to slow recently, but experts say it’s too early to tell if the trend will last.

On Friday, Dr. Rochelle Walensky, director of the US Centers for Disease Control and Prevention (CDC), said she is “cautiously optimistic” about the downward trend, but warned that the global case count continues to grow.

“The rate of increase is lower, but we are still seeing increases and of course we are a very diverse country and things are not uniform across the country. So we are looking at this with cautious optimism,” he said.

Last week there were an average of 337 new cases of monkeypox per day in the US, according to CDC data. This represents a decrease of 24% compared to the previous two weeks, a difference of more than 100 cases per day.

Some factors are “working together to turn the tide,” Walensky said, including vaccination, behavior changes and harm reduction messages “that are being heard and implemented.”

But there are many more factors that are still in motion, leaving unanswered questions.

Key factors to consider

Local health departments say they are working hard to understand the factors driving the case trends.

The city of Washington has been an epicenter of the outbreak from the beginning; Washington has had nearly 12 times as many cases per capita as the country as a whole.

The local health department has noted a “slight decline” in cases in the past week, but said “education and vaccinations” are two key factors that will drive trends going forward.

And “with 100,000 students returning to the city in the next week, it’s too early to project trends,” they said.

The California health department said trends in the state in recent weeks suggest “stabilizing rather than increasing levels of transmission” and daily cases “leveling off.”

“The trend could continue, but it will depend on vaccination and behavior change, as noted above,” they noted.

Dr. Ashwin Vasan, New York City’s health commissioner, said last week that the city has also noticed “cases starting to go down and transmission slowing down” in recent days.

Even if cases rise again, Donal Bisanzio, a senior epidemiologist at the nonprofit research group RTI International, predicts a “rapid decline” will occur at some point. Monkeypox “is a disease that offers total protection after you’ve caught it,” he said. Reinfections are not a factor.

But no one knows for sure what comes next.

“I think in all public health responses that I’ve worked on, we want a crystal ball. It’s very difficult to get to it early, even for things that we’ve understood much better for years,” said Janet Hamilton, director executive of the Council of State and Territorial Epidemiologists. “Tell me when flu season is going to end. I can’t.”

Multiple shoot edges

Experts say the ongoing outbreak is a rapidly evolving situation, making it difficult to know what to expect in the coming weeks and months.

“There’s a lot of interest in using models and thinking about what the trajectories would be under different scenarios,” Hamilton told CNN.

All data-driven models require that at least some reasonable assumptions be made. However, in the case of monkeypox, the United States is still collecting some of the critical pieces of “boots-on-the-ground epidemiology,” Hamilton said. It is a “highly infectious disease that is spreading very rapidly and that we don’t fully understand.”

Some of the stigma attached to the current outbreak may make people reluctant to seek medical care, discuss recent contacts or get vaccinated, experts say.

“I’m always concerned about stigma and discrimination because it can make it difficult to control an outbreak like this,” said Dr. Wafaa El-Sadr, a Columbia University professor of epidemiology who has researched HIV, emerging infections and other diseases. This could mean that people are not “taking advantage of the important tools at hand to stop this outbreak.”

Typically, only 15 to 25% of contacts are identified, which can skew understanding of the pace of disease spread, Hamilton said. “That ends up being a very different challenge that would then have a direct impact on the look of those models.”

Although men who have sex with men have been disproportionately affected in the current outbreak, anyone can contract monkeypox and an increasing number of recently reported cases have been among women and children, which could further change the transmission map.

And a recent change in the federal vaccination strategy, which potentially quintuples existing vaccine supply but poses other challenges, marks a major shift in another key forecasting factor.

More than 1 million vials of the Jynneos vaccine have been allocated to states and other local jurisdictions, according to data from the US Department of Health and Human Services. However, only about 208,000 doses have been reported administered to CDC.

In June, researchers at RTI International published a worked estimating the potential burden and duration of the monkeypox outbreak that was not yet endemic. They projected that every three cases could cause 18 secondary cases if prevention measures were not applied, and that the outbreaks would last between six and nine months. According to the model, contact tracing and vaccination of close contacts could reduce that risk by more than 70%.

But with new insights and information they tweaked the model and ran it again, with updated estimates due out in a few weeks.

“We’re talking about a model that was done right in the first month of transmission out of Africa. We thought everything was going well,” said Bisanzio, who is a co-author of the RTI report. The model assumed that the public health response would be quick and effective, especially after the covid-19 pandemic.

“We know that this story then played out differently,” he said, with a delay in response time, doubts about vaccination and concentrated transmission in a high-risk community that no one expected. “I can say that the figures (in the updated model) are much higher than what we estimated at the beginning.”

That said, another major tweak to his model comes from a recent survey conducted on men who have sex with men, in which nearly half of those surveyed said they had changed their behaviors to protect themselves amid the ongoing outbreak.

The Institute for Health Metrics and Evaluation (IHME) at the University of Washington built an influential forecasting model for Covid-19 early in the pandemic, but it won’t do the same for monkeypox.

“In our view, this disease is not going to spread as widely in the community as we see with COVID,” said Ali Mokdad, an epidemiologist and professor of health metrics sciences at IHME.

There are some similarities to HIV that are concerning, but “we don’t see that playing out here,” he said, since monkeypox leaves the body much faster and is much less deadly.

Although models based on extensive data may be largely absent, experts say they are watching trends closely.

“It doesn’t take a model to know that right now the strategy should be to identify high-risk individuals and vaccinate as many as possible,” Hamilton said.

— Jacqueline Howard contributed to this report.