ZIP code data for COVID-19 cases still comes with questions in Arizona
PHOENIX — So it turns out that some areas of the state, like the near south side of Tucson, may not be quite the COVID-19 hotspots that newly released data would show.
The numbers released by the Department of Health Services on Sunday show the 85714 ZIP code with 64 confirmed cases. By Monday it was 67.
That ZIP code has one of the highest case counts in the state. The Peoria ZIP code of 85382 has the most with 69 as of Monday morning.
But here’s the thing.
State health officials report that if they don’t know the home address of the patient, they include it in the ZIP code of the medical provider, followed by the address of the facility that reported the case.
And what’s not clear is how many of the cases listed for 85714 — the area between Ajo Way and Irvington Road — are simply patients at Banner University Medical Center South.
The fact that the 85714 data may include the hospital does not necessarily mean there isn’t a hotspot on Tucson’s south side.
The 85706 ZIP code just south, down to Drexel Way, reports 49 cases.
But there are some indications the data are skewed.
Consider: the 85714 ZIP code, according to most recent census data, has 15,138 residents. So the rate of infection would be about 4.5 per 1,000 persons.
By contrast, the 49 cases in 85706, with 58,266 residents, comes out to 0.8 cases per 1,000 residents.
And in the area to the west of both ZIP codes, the 32 cases in 85746 translates of 0.7 cases per thousand.
In the Phoenix area, the 85008 ZIP code with Valleywise Medical Center — the old Maricopa Medical Center — shows 28 cases, versus from just 6-10 in the ZIP code just to the west. When population is figured in, that comes out to less than 0.5 cases per thousand.
And the rate is just a fraction of that in the 85006 ZIP code with Banner University Medical Center in Phoenix where fewer than 10 cases were reported,
Questions about hospital data aside, there are other gaps in the data.
Results for ZIP codes where Native Americans constitute a majority of the population were not released, though the health department provided no specific reason for suppressing the data other than they were waiting for tribal approval. But the statewide totals, including the fact that 15 percent of the 122 deaths as of Monday due to COVID-19 were of Native Americans, appears to incorporate those missing numbers.
Still, there are some data points that appear to support the disproportionate affect on Native Americans.
In the 86047 ZIP code, the area around Winslow, the 48 cases translates out to about 2.3 per thousand. And with 52 cases, the 86040 ZIP code in and around Page comes out at about 4.7 cases per thousand residents, a figure higher than even the ZIP code around Tucson’s Banner Hospital.
That, however, may not be true statewide.
Most notably, there are no reports of COVID-19 cases from Fort Huachuca in Sierra Vista. State officials had no immediate answers for the gap — and whether the 16 cases reported in Cochise County and 3,702 statewide include any military living on the base.
Ditto Luke Air Force Base in Glendale, where officials in March confirmed at least two cases.
There did appear to be some data from Davis-Monthan Air Force Base. But the ZIP code for that base listed five or fewer cases, versus more in adjoining ZIP codes.
Even some of the statewide data appears to lack specifics.
The health department said it did not know the race or ethnicity of 63 percent of those with a confirmed diagnosis and 49 percent of those who actually died.
The limits of the data drew a warning from Pima County health officials.
In a release Monday the agency pointed out that the information provided by the state reflects only individuals who have tested positive for the virus. But there also have been limits on how many people can get tested.
“Most tests so far have only been done for those who are very sick, are known contacts of an already confirmed case, or are healthcare workers,” the county release says. And that means the map released by the heath department shows where concentrations of people who got tested live — to the extent the data includes home address — not necessarily where the chances of contracting the virus are higher.
State Health Director Cara Christ, in a blog post, acknowledged as much.
“While physical distancing is occurring, people do not always remain isolated within their own ZIP code when conducting essential business,” she wrote.
“Finally, ZIP code counts include anyone in that ZIP code who has tested positive, whether their infection occurred two months ago and has resolved or two weeks ago and is still active,” Christ said. “Thus, was ZIP code data is informative it (ITALICS) should not (ROMAN) be used to determine whether or not residents of a particular ZIP code are or are not at risk of COVID-19.”