It’s not just testing — coronavirus spike can’t be attributed to expanded testing capacity, data shows

As coronavirus case counts pile up across the U.S., President Donald Trump continues to blame testing — in rallies, interviews, and on Twitter — for the recent surge in the outbreak. 

“If we didn’t test, you wouldn’t be able to show that chart,” Trump said in response to a question from Chris Wallace about rising U.S. Covid-19 cases in a Fox News interview that aired on July 19. “If we tested half as much, those numbers would be down.”

But a CNBC analysis of testing data found that even as the U.S. has increased its testing capacity, cases of the virus are being found at a higher rate, a pattern that contradicts what epidemiologists say should be happening as a country gets a pandemic under control.

“That claim is patently false,” said Dr. Yonatan Grad, a professor of immunology and infectious diseases at Harvard, in response to the idea that increased testing explains the recent outbreak. “It is at best misleading, and at worst intentionally subverting public health responses.”

In interviews with CNBC, epidemiologists pointed to the “percentage of positive tests” as a way to understand whether the growth in U.S. coronavirus cases is due solely to increased testing. When coronavirus tests are in short supply, only the sickest individuals are typically tested, causing the share of positive tests out of total tests to be high. But as more tests become available, those with mild or no symptoms — who are less likely to have Covid-19 — are able to get tests, which would lead to a lower positivity rate if the virus were not spreading.

The percentage of positive tests in the U.S. has increased from 5.4% on Memorial Day to 8.6% on July 23, according to a CNBC analysis of data from the Covid Tracking Project. Daily testing nationwide has nearly doubled over that period, from an average of 410,000 daily tests performed on May 25 to more than 775,000 daily tests as of July 23. To account for daily reporting fluctuations, CNBC’s analysis used a seven-day average of cases and tests to calculate percent positive rates.

The World Health Organization has said that in countries that have conducted extensive testing for Covid-19, the percent positive rate should remain at 5% or lower for at least 14 days.

“If the disease was not spreading and you were increasing testing, then the fraction positive should stay stable or go down,” Dr. Grad told CNBC. “But in fact we’re seeing that the fraction of tests that are positive is going up as testing is going up. That is a clear indication that there is increasing spread of the virus.”

Twenty-nine states have seen an increase in their percent positive rates since Memorial Day, and this trend holds true for the states experiencing the largest virus outbreaks. In each of the ten states that have seen the greatest increase in coronavirus cases between May 25 and July 23, the percent of positive tests has also increased over that period.

Of these ten states, none has seen more growth in its percent positive rate than Arizona. On May 25, the state’s rate of positive tests was 7.6%, a figure that sits at 25% as of July 23. Over that period, Arizona reported more than 136,000 new coronavirus cases, according to Covid Tracking Project data.

In California, more than 331,000 cases have been reported since Memorial Day, as the state’s percent positive rate has doubled from 4% to 8%. Dr. Julie Vaishampayan, the health officer in Stanislaus County east of San Jose, told CNBC that her area had been hit particularly hard in recent weeks, and that attributing the increase in reported coronavirus cases to more testing does not tell the full story of the outbreak.

“The more testing you have, the more you can actually identify cases,” said Dr. Vaishampayan when asked whether an increase in testing is the cause for increasing case counts. “But that’s not the whole truth.”

Dr. Vaishampayan said that as her county has expanded its testing capacity, it has also seen its percent positive rate increase from 3% to 20%. “We’re doing more testing, but finding way more positives,” she said.

The percent positive rate can be influenced by the types of populations tested. If existing testing resources were dedicated to at-risk populations, such as nursing homes experiencing virus outbreaks, a higher share of positives to tests would likely be recorded. But in Stanislaus County, at least, Dr. Vaishampayan said that they “haven’t changed anything” about testing sites that would naturally influence the percent positive rate, and she believes the same is true at the state level.

Epidemiologists also stressed the importance of looking at multiple metrics to track the spread of Covid-19, including hospitalizations and deaths. And testing, even if it does in part contribute to higher official case tallies, is vital to the Covid-19 public health response.

“More testing is good,” said Dr. Lorna Thorpe, a professor of epidemiology at New York University. “From a public health standpoint, the more cases we know about, the better. We want to know about cases so they can be treated and isolated and try and stop transmission.”