Reported coronavirus-related deaths appear to be on the rise in Florida, Texas, California, Arizona and some other states that are struggling to contain rapidly expanding outbreaks, a CNBC analysis of data collected by Johns Hopkins University shows.
After peaking at an average of more than 2,000 deaths per day just three months ago, primarily driven by New York and New Jersey, fatalities in the U.S. have been slowly declining — falling to an average of less than 600 fatalities a day from June 23 through July 8. Covid-19 deaths in the U.S. have declined or remained relatively stable for weeks, even though cases have more than doubled since mid-May. But the daily death toll appears to be on the rise again in the U.S., epidemiologists say.
Covid-19 fatalities have steadily ticked up across the nation with the average number of fatalities a day rising over the last three straight days to over 600 on July 9, based on a seven-day average of daily reported deaths, driven by surges in several hot spots. Epidemiologists say it is cause for concern that deaths are beginning to accelerate again, even if it's just a few days of data.
U.S. officials and the general public should have seen the rise in deaths coming, Larry Levitt, executive vice president for health policy at the Kaiser Family Foundation, told CNBC. Deaths tend to lag new cases because it can take weeks for a patient to get sick enough to be hospitalized and eventually die.
"This was predictable. We seem to have had difficulty in this country looking a few weeks in advance," Levitt said. "But we know the pattern that as more people get infected, more people get hospitalized and ultimately more people die."
Florida, Texas, California and Arizona have all seen their daily death tolls rise to record highs over the past three days, according to Hopkins data.
California has reported an average of about 85 new coronavirus-related deaths per day over the past seven days as of Thursday, up more than 29% compared with a week ago, according to CNBC's analysis of data compiled by Hopkins. The state's Covid-19 death toll now stands at 6,859, according to Hopkins.
Florida has recorded an average of 56 deaths per day over the past seven days, up over 35% compared with a week ago, CNBC's analysis shows. Hopkins' data shows more than 4,000 people have died of the disease in the state so far.
On Thursday, Texas reported an average of about 66 new deaths per day over the past seven days, up more than 106% over the past week, according to CNBC's analysis. More than 3,000 people have died of Covid-19 in the state so far, according to Hopkins.
To be sure, the fatality data is imperfect, epidemiologists say. If a Covid-19 patient has an underlying ailment, such as heart disease, and the virus worsens their condition and they die, the doctor can categorize cause as either. Elderly patients who die in nursing homes often have the coronavirus but aren't often tested, they've said.
"Record keeping can be all over," said Dr. Bruce Y. Lee, a professor of health policy and management at the City University of New York.
The country, however, is much better equipped today to handle an influx of Covid-19 patients than it was at the beginning of the outbreak, epidemiologists said. That should help avoid the same kind of spike in fatalities that overwhelmed hospitals and funeral homes in the Northeast and Washington state in March and April. Nonetheless, three epidemiologists in Florida and Texas all said they expect deaths to continue to rise for at least a few weeks.
"Our cases started to increase right around the beginning of June and now as I'm looking through, you can see that the deaths have started to trend upward a little bit as well," said Cindy Prins, an epidemiologist at the University of Florida. "Initially, a lot of people were saying, well, it's flat, it's flat. And the concern there was, well, we haven't caught up with data, and now we are starting to see that increase, which is definitely a concern."
State officials in Florida and other states have noted that the recent surge in cases is driven largely by younger patients, which is significant because young people are less likely to become severely sick and die from Covid-19. However, the Covid-19 data shows that those infections are increasingly spreading to older, more vulnerable people, which could be driving the uptick in deaths, Prins said.
"There is more testing now than there was then, so that may account for some of this, but I think we're seeing a true increase in cases in older adults, which makes sense given the overall large increase in cases," she added.
The shift from younger people to older people is beginning to show up in the data, said Dr. Mary Jo Trepka, an epidemiologist at Florida International University. Last month, the state reported that the daily median age of newly diagnosed Covid-19 patients hit a record low of 33. On Thursday, however, the median age of newly infected people had risen to 40, according to the state's health department.
With that median age ticking upward, both Prins and Trepka said they expect deaths to continue to rise in the coming weeks. However, Trepka noted that deaths won't likely rise at the same rapid pace as New York City, which was hit particularly hard early on in the U.S. pandemic. Public health officials have since instituted measures to protect vulnerable populations and hospitals have improved patient care since then.
"It doesn't appear to be the same rates as back in April, and I think health care has dramatically improved. Care providers are much more skilled at caring for people with Covid-19," she said. "Nevertheless, with these large numbers of cases, I do think that we're going to be seeing continuously more deaths."
Deaths caused by Covid-19 began to increase slightly in Texas about two weeks ago, according to Spencer Fox, associate director of the University of Texas-Austin Covid-19 Modeling Consortium.
"I don't think it's anything unexpected," he said in an interview with CNBC. "I think it was more so a question of when we would start seeing an uptick, rather than if we would start seeing an uptick."
His team's model does not predict as rapid an increase in deaths as was seen in March and April in the Northeast and some other parts of the country, he said. But hospitalizations have risen at a worrying pace, he said, indicating that older and more vulnerable people are getting infected. He added that infections in younger people was a "leading indicator" of a worsening outbreak that was bound to affect the more vulnerable populations in the state.
"This resurgence might have started in younger populations; maybe they were the first to be infected. But clearly those populations aren't insulated from older individuals," Fox said. "This is a real resurgence in the epidemic. It's not limited to just younger individuals who are more likely to survive it. It's everywhere."
He added that his team's model predicts that deaths will continue to increase for two weeks "at least, if not longer, depending on really how the state reacts."
It's difficult to get an accurate understanding of the reality of the outbreak by looking only at the national numbers, Kaiser's Levitt said, because the progress places like New York have made in combating the outbreak offsets the worrying numbers elsewhere. He added that the death toll is an especially difficult figure to track because of differences in reporting standards across states.
He said now that there's an observable increase in deaths, the trend is likely to continue for a number of weeks or even months as people who recently got infected fall ill, get hospitalized and eventually die.
"I think in the next week, the pattern of increasing deaths is going to become clear," he said. "And it will no longer be possible to claim that the declining mortality is somehow a success."