Unannounced 'patient-zero' drills across 73 New York-area healthcare facilities found that 40% failed to properly mask and isolate a symptomatic actor, with only 25% of clinicians wearing all recommended PPE, raising concerns about hospital readiness for a bird flu or respiratory virus outbreak.
<p>COVID-19 made the weaknesses in our pandemic preparedness at national and global levels painfully clear. But a new study highlights that even at the first, local steps of an outbreak with pandemic-potential, we are worryingly underprepared.</p>
<p>Health officials in New York set up a regional test of healthcare facilities to identify and respond to a patient with a mystery infection that could pose a pandemic-level threat. But things didn't go well. Overall, responses were slow, appropriate use of personal protective equipment was spotty, and only 60 percent of facilities passed the basics of the test—an overall failing grade.</p>
<p>"These drills identified needs for health care workers and health care facility staff member training to achieve earlier identification, masking, and isolation of potentially infectious febrile patients," the officials wrote in the study, which was <a href="https://www.cdc.gov/mmwr/volumes/75/wr/mm7538a1.htm?s_cid=OS_mm7538a1_w">published in the Morbidity and Mortality Weekly Report</a>.</p><p><a href="https://arstechnica.com/health/2026/10/patient-zero-drill-put-health-facilities-to-the-test-40-of-them-failed/">Read full article</a></p>
<p><a href="https://arstechnica.com/health/2026/10/patient-zero-drill-put-health-facilities-to-the-test-40-of-them-failed/#comments">Comments</a></p>
# Patient-zero drill put health facilities to the test—40% of them failed
Source: [https://arstechnica.com/health/2026/10/patient-zero-drill-put-health-facilities-to-the-test-40-of-them-failed/](https://arstechnica.com/health/2026/10/patient-zero-drill-put-health-facilities-to-the-test-40-of-them-failed/)
In the drill, which ran between January and June 2026, actors visited 73 healthcare facilities\. Most were in New York City \(52\), with 15 in New York outside the city, three in New Jersey, and three in the US Virgin Islands \(which the US considers part of the same[health region](https://www.hhs.gov/about/agencies/iea/regional-offices/region-2/index.html)\)\.
Actors mostly visited emergency departments \(42 of 73\), but also hospital outpatient clinics \(19 of 73\), and urgent care centers \(12 of 73\)\. The facilities had agreed to be part of the drills but didn’t know when they would occur\.
## Failing grades
Overall, only 60 percent \(44 of 73\) of the facilities both masked and isolated the actors\. Only about 55 percent of emergency departments did both, while 68 percent of clinics and 67 percent of urgent cares passed\. Eighteen percent of facilities didn’t offer a mask to the coughing patient who said they had a fever\.
Of those that masked and isolated, fewer than half \(43 percent\) did so within the one\-minute goal, and less than half \(48 percent\) moved the patient to an isolation room within the 10\-minute goal\. Outpatient clinics had a median time to isolation of 20 minutes, with at least one clinic taking nearly an hour\. Overall, the median time to mask was 2 minutes, and the median time to isolation was 11 minutes\.
Most of the healthcare facilities \(93 percent\) screened the actor for symptoms\. But—despite the dead duck tale—staff at only seven of the 73 facilities asked questions specific to bird flu exposure\.
Clinicians interacting with the actors inconsistently wore recommended personal protective equipment \(gloves, mask, gown, and eye protection\)\. Only 25 percent wore all the recommended PPE; 6 percent wore none\. Nineteen percent of clinicians didn’t wear a mask or respirator\. Staff at only 40 of the 73 facilities \(55 percent\) notified or planned to notify internal Infection Prevention and Control \(IPC\) staff as recommended\.
The poor pass rate and the lengthy times spell trouble\. “Health care facility waiting areas can be high\-risk settings for transmission of respiratory viruses if appropriate IPC measures are not implemented,” the authors highlight\.
These substandard results may also overestimate preparedness\. Not all the healthcare facilities invited to participate in the drills agreed to do them—and those facilities that declined may have scored even worse\. Also, although the drills were unannounced, some facilities may have had a heads\-up, which could have boosted their results\.
Quest Diagnostics suffered a breach of its Quanum LSM lab platform, illustrating a recurring pattern of centralized healthcare systems exposing millions of patient records. The post cites Synnovis, HHS breach statistics, and IBM cost data to argue for keeping patient data decentralized at the hospital level.
A third US person has died of bird flu (HPAI H5) since the virus jumped from wild birds to dairy cows, with the latest fatality linked to exposure at an infected Washington dairy farm. Health officials say three cases have been reported in Washington's Yakima County, though risk to the general public remains low.
A GP trainee built a medical consultation game (doctorfoo.ai) and benchmarked 13 AI models across 195 consultations: all models correctly diagnosed every case, but safety behavior (catching red flags like contraindicated prescriptions) varied widely, with GPT-6, GPT-6.1 Sol, and Claude Opus 5.5 performing best and price barely predicting quality.
Healthcare workers at HCA Healthcare report that the Palantir-co-developed AI scheduling tool Timpani, rolled out across ~130 hospitals, ignores nurse shift preferences and creates understaffed, inexperienced care teams, raising patient safety concerns.