Alaska does more COVID-19 tests with fewer positives than most other states. That doesn’t tell the whole story.

Alaska conducts more COVID-19 tests per capita than almost any other state in the country, a fact Gov. Mike Dunleavy mentions often at community briefings.

Alaska’s testing positivity rate — the number of positive tests out of total tests performed — has also remained far below the national average, even though it’s rising slightly as time goes on. The state’s death rate among people with coronavirus is the nation’s second lowest.

Yet the state’s new COVID-19 case counts in communities like Anchorage and Fairbanks are soaring right now. New cases, especially among younger people, prompted health officials to warn of significant community transmission around the state.

Do the testing numbers tell the whole story? Not entirely.

Comparing Alaska to other states is complicated by the state’s unique strategy to screen travelers and seafood workers that adds a large pool of healthy people less likely to test positive for the virus.

Alaska’s unique geographic isolation and fragile health care resources also prompted state health officials to adopt an aggressive testing strategy early on.

The positivity numbers are promising indicators the state is generally doing enough testing, officials say.

But they don’t account for the dramatic increases showing up in new daily cases.

“We do believe that we’ve got increasing community spread, not just increasing testing,” Dr. Anne Zink, the state’s chief medical officer, said this week.

In the top 3

By Friday, Alaska was the third most-tested state in the nation. Only New York and Louisiana test more people per capita.

According to a national data initiative run by Johns Hopkins School of Medicine, Alaska currently tests about a quarter of its total population — or 21,167 tests for every 100,000 residents – which is more tests per capita than all but those two states.

Widespread testing is crucial to gauge the spread of COVID-19 and to catch potentially contagious infected people with few if any symptoms, health officials say.

The current “best guess” is that about 40% of COVID-19 infections involve people without symptoms, said Dr. Jay Butler, Alaska’s former chief medical officer who is now deputy director for infectious diseases at the federal Centers for Disease Control and Prevention.

“That number may very well change as we learn more about this disease,” Butler said in an interview from Anchorage, where he was working remotely for a week after helping head the national COVID-19 response. “Also, there’s a lot of very mild disease that occurs that may not be detected apart from very aggressive testing.”

Alaska’s testing is certainly aggressive when taken as a whole. But comparing our testing rates to that of other states is complicated.

For one, the state adopted an unusually broad testing policy almost from the start of the pandemic to identify cases early and minimize the odds that a surge would overwhelm hospitals with limited resources, officials say.

One-fifth of acute care hospital beds in Alaska are in hospitals off the road system, in difficult-to-reach parts of the state, according to the Alaska State Hospital and Nursing Home Association. Alaska’s ratio of hospital beds to people is on the low end of the spectrum. It can also be a struggle just getting testing swabs to rural villages.

Alaska struggled initially but then took steps to ramp up testing capacity faster than many other states given our small population size. That sets our rates apart.

“It does make it hard to compare our state with other states,” Zink said during a science briefing Wednesday. “But it’s also hard to compare our hospital capacity with other states.”

‘Tricky’

Another factor that really makes Alaska’s numbers hard to compare is the large number of relatively healthy people tested this summer due to screening mandates for travelers and seafood workers. That makes for a higher number of tests, but probably also more people with negative results because they’re not sick.

Under a state health mandate that went into effect in June, travelers into Alaska must either show proof of a negative test, get tested at the airport or agree to quarantine.

In five weeks, Alaska screened more than 78,000 people as part of airport testing, Zink said on social media earlier this week. More than 27,000 tested before traveling, nearly 27,000 were tested at an airport, more than 8,800 chose to quarantine and more than 16,000 were involved in a screened industry or a previous COVID-19 case.

There were a total of just 134 positive test results.

Seafood industry and other workers are also required to get tested multiple times during their time in Alaska. In March, state health officials estimated that industry would need between 140,000 and 160,000 tests this summer.

So far, the Alaska Department of Health and Social Services has distributed about 65,000 testing kits, said state pharmacist Coleman Cutchins. The industry used most of them, Cutchins said.

State testing data also lumps together Alaska residents tested out of state, nonresidents in Alaska, travelers and seafood workers getting tested when they arrive, according to Dr. Louisa Castrodale, the state’s infectious disease program manager. But the state isn’t able to separate percent positive rates in residents and nonresidents, who may be skewing the numbers.

“Those are tricky things to try and figure out,” Castrodale said.

The numbers

Coronavirus cases are spiking in Alaska, especially in Anchorage and Fairbanks, due to a combination of increased travel and group activities including churches, residential living facilities, workplaces, bars and social gatherings, state officials say.

Young people — who officials say may be more likely to socialize and go to work, and less likely to mask up or social distance — are driving the uptick: Most recent new cases have been people in their 20s and 30s.

A state update for the week reported more than a fifth of all the infections in Alaskans since the pandemic began were discovered last week, “suggesting a large increase in transmission in the state.”

Generally, the state’s worst indicators remain relatively low. Seventeen Alaskans have died with the virus.

A total of 96 residents have been hospitalized since March, giving the state a relatively good hospitalization rate. But early this week, physicians reported an increase in COVID-19 patients showing up at emergency rooms, a potential indicator for future hospitalizations. Nineteen people were hospitalized with the virus in Anchorage on Friday and another 13 patients were categorized as under investigation.

The soaring case numbers have overwhelmed contact tracing capacity, at least in Anchorage, foiling an essential part of the state’s strategy to contain the spread of the virus.

State officials say any gathering, especially indoor ones, poses an exposure risk. They urge people to keep their social circles small, wear face coverings and avoid large groups unless they can stay at least 6 feet apart — ideally outside.

A crude measure

Despite the state’s soaring case counts, Alaska’s positivity rate late this week was 1.78% over a seven-day average, the seventh lowest in the country.

Health authorities track testing positivity rate because it’s an important indicator of whether a community or state is doing enough testing to find coronavirus infections and stop the spread of the virus. The World Health Organization has said countries with broad testing should have a positivity rate that stays below 5% for 14 days.

A high positivity rate — Fairbanks hit 14% this week — can indicate that only the sickest people are getting tested and milder cases or asymptomatic cases are getting missed.

A lower positivity — Alaska’s overall rate this week was 1.85% and Anchorage was at 2.2% — can indicate a community is including people with mild or no symptoms.

The national seven-day moving average was at 8.5% positive tests as of Thursday, according to the Johns Hopkins Coronavirus Resource Center.

Public health officials have said that more testing is a good thing, because it allows for a more complete picture of where and how the virus is spreading. But it doesn’t tell the whole story.

Looking at percent positive can be useful, because it can tell you if you’re testing enough of the population, said Tom Hennessy, an infectious disease epidemiologist at the University of Alaska.

A positive rate of over 10% indicates that you may not be testing enough, he said, citing the World Health Organization.

“It doesn’t tell us if testing is getting to all the people who need it, or if there are populations that aren’t getting tested,” he said.

That is why looking at an overall testing positivity rate is just “a crude measure,” he explained.

The state plans to add to the COVID-19 data dashboard a metric that breaks down the percentage of positive cases by region, according to Zink. The more specific the data, the easier it can be to spot trends.

This week, officials said, they noticed that Fairbanks’ percent positive rate had “creeped up to 14%,” even though the state’s 14-day average was just under 2%.

The high rate prompted state epidemiologist Joe McLaughlin and others to meet with Fairbanks leaders to discuss testing there “to make sure that we’re doing a good enough job,” he said.

That’s another story Alaska’s numbers don’t tell.

Correction: An earlier version of this story incorrectly reported Thomas Hennessy’s job title. He retired from his role as head of the Arctic Investigations Program in Anchorage, and is now an infectious disease epidemiologist at the University of Alaska.

[Related: Rapid spread of coronavirus in Fairbanks prompts area mayors to encourage use of masks]

[Related: Beyond Anchorage, most of Alaska declines to mandate COVID-19 masks]

[Related: Anchorage officials warn of possible new restrictions amid ‘steep cliff’ of COVID-19 cases and ICU concerns]