Skip to content

New data shows COVID-19 deadlier, more prevalent in Pa. long-term care than once thought

Staff at Cedarbrook Senior Care & Rehabilitation in South Whitehall Township at a Healthcare Heroes Parade in May 2020, the height of the coronavirus pandemic.
Amy Shortell / The Morning Call
Staff at Cedarbrook Senior Care & Rehabilitation in South Whitehall Township at a Healthcare Heroes Parade in May 2020, the height of the coronavirus pandemic.
AuthorAuthor

The Pennsylvania Health Department has long alleged underreporting of COVID-19 in health care workers, but a recent update in test processing found nearly 31,000 more cases of the virus in long-term care employees than previously believed.

The update also discovered nearly 13,000 fewer long-term care residents tested positive, meaning the virus has proved to be more deadly among those residents, who are often medically fragile and at high risk for virus complications.

The sudden and dramatic shift in cases surprised industry representatives and senior advocates, who have complained about the discrepancies and omissions since the state began posting facility data following public demands for transparency. They argue accurate data is crucial to developing effective public health responses.

“We have been confused about data since they started reporting in May 2020. We have been asking the state to do better all along,” said Diane Menio, executive director of the Center for Advocacy for the Rights and Interests of the Elderly, a Philadelphia advocacy group.

The Pennsylvania Department of Health last month quietly released the revised testing data. A spokeswoman called the changes part of the department’s ongoing efforts to refine and improve reporting accuracy.

Among those efforts was the deployment of a new algorithm for categorizing positive residents and employee cases in long-term care facilities, spokesperson Maggi Barton said. The algorithm was launched Nov. 15 and applied retroactively.

As a result, positive cases among nursing home residents dropped from 79,904 to 66,911 between Nov. 10 and Nov. 14, and positive cases among facility staff nearly tripled from 17,453 to 48,256, according to the state.

As of Nov. 30, state health officials reported 68,315 positive cases among long-term care residents and 50,392 among employees. In the 19 days after the new algorithm was launched, 1,404 more residents and 2,136 employees tested positive for COVID statewide.

With fewer positive cases, the death rate among residents increased from 18% of those testing positive as of Nov. 10 to nearly 23% Nov. 30.

After steady and dramatic declines the first seven months of the year following the vaccine rollout, death rates among long-term care residents are rising, hitting 782 in November, the highest since February.

How the data changed

Throughout the pandemic, state health officials have faced challenges with collecting, analyzing and reporting the millions of test results it receives daily from health care providers and labs, Barton said.

Testing records and reports typically do not contain information about whether someone lives or works in a long-term care facility, which leaves it to the state to analyze and match positive cases with individual facilities, she said.

The matching was done using strategies and tools including self-reported outbreaks, case investigations, matching patient addresses to facility addresses and, most recently, tracking facilities ordering COVID-19 tests. Under a federal mandate, weekly testing is required for unvaccinated skilled nursing home employees as of Oct. 1.

The department also has acknowledged problems throughout the pandemic with facility administrators failing to regularly or accurately report cases and deaths daily to the state despite an order, which was rescinded at the end of October. The state now only uses the self-reported data the Centers for Disease Control and Prevention requires facilities submit to the National Healthcare Safety Network.

The improvements along with a new algorithm allowed the department to connect the addresses of the facilities receiving COVID testing supplies to the test results, Barton said. Previously, positive employee cases may have been attributed to a home address, if that was the only information the individual provided at the time of the testing.

As a result it was found that thousands of positive tests were incorrectly attributed to nursing home residents when they occurred in employees, and tens of thousands of positive results were linked to employees of long-term care facilities.

The Pennsylvania Health Care Association, which represents more than 400 long-term care operators in the state, said it was unaware the department is using a new algorithm that resulted in such a dramatic shift in positive case numbers.

But, it pointed out, state health officials have made it difficult for even lawmakers to get a complete look at COVID numbers.

Legislative Budget and Finance Committee members complained in a recently released report on COVID-19 death reporting that it received heavily redacted and incomplete data, according to PHCA President and CEO Zach Shamberg.

“Throughout the pandemic, we have expressed concerns over the management of state data and the surveys they have conducted,” Shamberg added.

At her organization’s last meeting with the Health Department, Menio asked that it use the Centers for Medicare and Medicaid Services database for tracking COVID-19 numbers in long-term care facilities.

“Although it’s not completely reliable, it’s better than what we have. I suggested they use that data for Pennsylvania reporting,” she added. “Now I see that they simply put a link to the data; not helpful.”

Menio added it should surprise no one that far more long-term care staff were infected with COVID than what data initially showed. She suspects the revised numbers are likely even higher.

“They were so unprepared for the pandemic, they made little progress on infection control, and their vaccination numbers are lower than for other health care providers,” she said.

Teresa Osborne, manager of advocacy and outreach for the Pennsylvania chapter of the AARP, also was unaware the state was using a new algorithm for categorizing cases in long-term care.

“While the ‘new’ data appears more correct, we have inquired with the department and await a response on the details on why the change was needed,” Osborne said.

Information about new positive cases in facilities remains elusive, according to some families, though safety precautions such as face masks and pre-visit questionnaires remain in place.

Newtown Township resident Lynne Engle said the Berks County memory care facility where her 91-year-old father lives provided COVID booster vaccines for residents and provides regular updates on his health.

“But they have not reported any information on the number of positive tests or new outbreaks at the facility,” she said.

As of Nov. 26, two residents and three staff at the facility tested positive over the previous two weeks, according to state-reported data. As of Nov. 26, 59 residents and 30 staff have tested positive for COVID, and nine residents have died.

As far as Mary Stackhouse knows, there have been no new COVID cases at the Northampton nursing home where her 88-year-old mother has been a resident since August. She heard those cases were in recently transferred short-term rehab patients.

As of Tuesday, the Holland Center for Rehabilitation has reported no new cases of COVID among staff or residents in the previous four weeks. A total of 50 residents — including Stackhouse’s mother — and 22 employees have tested positive and 16 resident deaths have been attributed to COVID since the start of the pandemic.

“There have been a few deaths since August. To the best of my knowledge, they were not attributed to COVID,” added Stackhouse, a Newtown Township resident. “This is all good. We certainly hope that remains the case.”