CDC MMWR: Modeled Scenario Projections for the Ebola Disease Outbreak Caused by Bundibugyo Virus, 2026

“All models are wrong, but some models are useful.” George E. P. Box (18 October 1919 – 28 March 2013) - Professor Emeritus of Statistics at the University of Wisconsin

#19,190

Late yesterday afternoon the CDC held an update (see transcript & video) on the the Ebola virus outbreak in Central Africa, focusing specifically on 3 new reports published Friday in the CDC's MMWR.  

• The first is an  Assessment of Risk to the U.S. Population from the Ebola Disease Outbreak Caused by Bundibugyo Virus, 2026 which finds `. . . the risk posed by this ongoing outbreak to the U.S. population during the next 3 months as low.'

• The second is Notes from the Field: Outbreak of Ebola Disease Caused by Bundibugyo Virus — Democratic Republic of the Congo and Uganda, May 2026 which reports `The scope of the outbreak is likely larger than that represented by available data and might prove challenging to contain and control.'

The third report, however - which modeled a variety of possible scenarios for the Ebola Outbreak - was the primary topic of discussion during this presentation.  

As with all such models, this isn't a prediction of what will be, only what might be, if certain things aren't done to prevent it. 

Based on the following comment from Dr. Pillai during the teleconference, the example given in the summary (70% of cases isolated in 1st 48 hours) doesn't currently appear to be happening:

Dr. Pillai

Currently, the situation is very fluid, and while the numbers are not completely known, based on the trajectory of the outbreak and the rapid extension into multiple different health zones over a short period of time, this appears to be in one of the lower end of the percentage of individuals that are being detected and isolated.

The range of outcomes - based heavily on success in identifying and isolating cases - is depicted in the following chart. If the success rate is < 50%, the risks of seeing 20,000+ cases over the next 3 months rises markedly. 

First, some excerpts from the report, after which I'll have a bit more. 

Modeled Scenario Projections for the Ebola Disease Outbreak Caused by Bundibugyo Virus, 2026
Early Release / June 5, 2026 / 75
 
Eric Q. Mooring, ScD1,2; William T. Koval, PhD1; Isobel Routledge, PhD3; Inga Holmdahl, PhD4; Guido España, PhD1; Rebecca Kahn, PhD4; Beau B. Bruce, MD, PhD1  

Summary

What is already known about this topic?

An outbreak of Bundibugyo virus disease (BVD), a type of Ebola disease, is currently ongoing, centered in the Ituri province of the Democratic Republic of the Congo (DRC).

What is added by this report?

CDC used a transmission model to project outbreak growth over 3 months, by using different assumptions about the number of deaths as of May 24, 2026, and by varying the percentages of persons with BVD who are successfully identified and isolated to prevent ongoing transmission. Assuming 50 cumulative deaths as of May 24, 2026, if 70% of patients were to enter isolation, only approximately one in 20 simulations projected an outbreak exceeding 10,000 cases within 3 months.

What are the implications for public health practice?

Large-scale, rapid public health action is needed to control the current outbreak, already the largest known BVD outbreak, from becoming one of the largest Ebola epidemics in history.

        (SNIP)

Outbreak Size Projections and Inferred Spillover Date by Assumed Number of Deaths

Assuming 50 deaths. The model calibrated to 50 deaths estimated that the spillover event that triggered this outbreak most likely occurred on approximately February 19, 2026 (interquartile interval [IQI] = February 1–March 8). Assuming that 20% of infected persons were successfully isolated beginning May 24, 2026, projections showed ≥20,000 cumulative cases in 65% of simulations, ≥10,000 cumulative cases in 85% of simulations, and ≥4,000 cumulative deaths in 69% of simulations (Figure).

Even with 50% of infected persons isolated, many simulations still projected these numbers of cases but were less likely to occur (17% of simulations projected ≥20,000 cases and 22% projected ≥4,000 deaths). At 70% isolation, projected outbreaks were much more likely to be smaller, but still of substantial size, with 94% of simulations projecting <10,000 cases and only 1% projecting ≥20,000 cases; similarly, at this isolation level, 90% of simulations projected <2,000 deaths and only 3% projected ≥4,000 deaths. Re declined proportional to the percentage of infected persons successfully isolated (Supplementary Figure 1).

Assuming 100 deaths. Assuming 100 cumulative deaths as of May 24, 2026, the inferred median spillover date was February 8, 2026 (IQI = January 21–February 27). Very large outbreaks were likely in the scenario in which only 20% of patients were isolated (76% of simulations projected ≥20,000 cases and 87% projected ≥4,000 deaths). In the scenario in which 70% of infected persons were isolated, 73% of simulations projected <2,000 cumulative deaths by August 22, 2026, and 10% projected ≥4,000 deaths (Supplementary Figure 2).

Assuming 200 deaths. Assuming 200 deaths by May 24, 2026, the calibrated model inferred a median spillover date of January 29, 2026 (IQI = January 9–February 18). The earlier spillover date would have generated a larger outbreak by the time interventions began; thus, even with 70% of infected persons isolated, 42% of simulations projected ≥10,000 cases by August 22, 2026.

Sensitivity to Basic Reproductive Number

Simulated outbreaks with R0 values higher than the median R0 typically reached ≥10,000 cumulative cases and ≥2,000 cumulative deaths by August 22, 2026, in scenarios with ≤50% isolation, even assuming only 50 cumulative deaths by May 24. In the scenario with 70% of infected persons isolated and 50 assumed deaths by May 24, 2026, no simulations projected ≥2,000 deaths when R0 values were lower than the median R0, but 20% of simulations projected ≥2,000 deaths when R0 values exceeded the median (Supplementary Figure 3).

        (SNIP)

The high probability of a large outbreak over a 3-month period primarily results from the large size of the outbreak at the time it was initially confirmed. This analysis did not provide evidence that R0 for this outbreak is unusually large.† Time between Ebola outbreak onset and detection is positively correlated with overall outbreak size and duration (4).

CDC’s assessment that the risk to the general U.S. population is low (5) is not changed by this analysis. Despite the unprecedented size of the 2014–2016 West Africa Ebola epidemic, only two Ebola transmission events occurred in the United States. Those two infected persons were health care workers caring for a patient with Ebola who had traveled to the United States before enhanced screening, risk assessment, and health education measures were implemented at U.S. ports of entry (6). Both persons infected in the United States recovered.

       (Continue . . . )
 

All three MMWR reports are worth reading in their entirety, and the video presentation (21 minutes) is very much worth watching. 

While conditions could change, right now the DRC and surrounding countries appear to be on a trajectory that could eventually equal or even exceed that seen during the 2014-2016 West African Ebola Outbreak (28K cases, 11K deaths). 

But, as they say, `If you've seen one epidemic . . . you've seen one epidemic.'  The Bundibugyo virus is a different threat than Ebola Zaire, and the conditions in the DRC, Uganda, and South Sudan differ as well. 

For now, this is 99% a regional threat.  Exported cases are a possibility, but large outbreaks in places like Europe or North America - which are far better prepared to deal with this virus - are unlikely. 

But all of this assumes that reasonable containment efforts are made - or even possible - at ground zero.  Long chains of human-to-human transmission are problematic with any zoonotic virus, as it increases the chances that the virus will better adapt to a human host.

Which makes it very much worth our while to do whatever we can to help  bring this outbreak under control.

Sooner, rather than later. 

FluView Week #20: EOY Review of Increased Detection of Oseltamivir Resistant H1N1 Viruses

Abrupt rise in Oseltamivir resistance in Catalonia, Spain - fall 2025

#19,189

For the past 3 years we've been following sporadic reports of `reduced susceptibility' of the seasonal H1N1 flu virus to the antiviral drug oseltamivir (aka `Tamiflu') around the globe (see EID Journal: Multicountry Spread of Influenza A(H1N1)pdm09 Viruses with Reduced Oseltamivir Inhibition, May 2023–February 2024).

While concerning, these reports indicated reduced inhibition - not complete failure - and its incidence has been modest; typically in the low single digits.   

The main culprit in all of this has been an increase in the S247N mutation (often combined with I223V), which can reduce the effectiveness of oseltamivir 7 to 16 fold. 

Less common, but far more impactful, is the H275Y mutation, which can effectively render oseltamivir useless. This was the cause of the `Tamiflu failure' of 2008, which temporarily forced the CDC to change their flu treatment guidelines.

Luckily, the arrival of a new, and still susceptible pandemic H1N1 virus in 2009 granted oseltamivir an unexpected reprieve, but since then we've kept a close eye out for any genetic changes that could affect its effectiveness. 

Last summer, in Virus Research: A 15-year Study of Neuraminidase Mutations and the Increasing of S247N Mutation in Spain, we looked at a study that found a sharp increase in detections of the S247N mutation beginning in 2024. 

Highlights

• In a landscape of a very narrow arsenal of influenza antivirals, resistance mutations are a significant threat.

• Resistance mutations were present in 0.5-5% in A and B influenza viruses during the last 15 years.

• However, S247N resistance mutation in the NA gene sharply increased during 2023-2024 season.

• While this mutation does not confer strong resistance by itself, their fixation could increase the risk of resistance in the future if other resistance mutations appears or get fixed together with it

Over the 2025-2026 flu season we've continued to see scattered reports of resistance, including Taiwan's CDC reported that 6.5% of the H1N1 viruses characterized in 2025 showed signs of oseltamivir resistance, and China: National Influenza Center Reporting Increased Oseltamivir Resistance in Seasonal H1N1. 

As for detections in the United States, as we saw last February, over the entire 2024-2025 flu season - out of 1697 H1N1 viruses tested - only one carried the NA-I223V and NA-S247N amino acid substitutions.

And reassuringly, during the first 13 weeks of the 2025-2026 flu season (Oct - Dec) the CDC reported zero elevated resistance among the first 193 H1N1 viruses tested (see screenshot from FluView Wk 53).  

But by the end of February, 2026 (FluView Week 7), things had begun to change. Based now on 517 H1N1 isolates tested since October 2026 - the CDC reported 10 viruses with reduced susceptibility and 4 with highly reduced inhibition.

Four A(H1N1)pdm09 viruses had NA-H275Y amino acid substitution conferring highly reduced inhibition by oseltamivir and peramivir. Ten A(H1N1)pdm09 viruses had amino acid substitutions NA-I223V and NA-S247N and showed reduced inhibition by oseltamivir. Two B viruses had amino acid substitution NA- M464T and showed reduced inhibition by peramivir.

Last Friday the CDC published their last full FluView report of the 2025-2026 flu season (week 20), which contained the following brief update on oseltamivir resistant viruses. 

Nine A(H1N1)pdm09 viruses had NA-H275Y amino acid substitution conferring highly reduced inhibition by oseltamivir and peramivir. Nineteen A(H1N1)pdm09 viruses had amino acid substitutions NA-I223V and NA-S247N and showed reduced inhibition by oseltamivir. One A(H1N1)pdm09 virus had amino acid substitutions NA-I223T and NA-S247N and showed reduced inhibition by oseltamivir. Two A(H3N2) viruses had amino acid substitution NA-E119V conferring highly reduced inhibition by oseltamivir. Three B viruses had amino acid substitution NA-M464T and showed reduced inhibition by peramivir.

While the number of S247N+I223V mutations remains low (2.1%), this is more than a 35-fold increase over the previous year, making this a trend well worth following. 

The incidence of the far more impactful H275Y mutation (1%) remains about average, but there are concerns that the stacking of H275Y with either I223V or S247N could greatly enhance its impact.

None of this is to suggest we are on the verge of another antiviral crisis, only that we continue to see some concerning trends. And given the gaps in global influenza surveillance and reporting, we are only seeing part of the picture.

As we are so frequently reminded - evolution never stops - and while our current antiviral armamentarium remains effective against seasonal H1N1, the events of 2008 reminds us how quickly that can change. 

Stay tuned.

USDA: USDA Confirms Presence of New World Screwworm in the United States

 

#19,188

The New World Screwworm (NWS aka Cochliomyia hominivorax) - which was eliminated in the United States during the 1960s - is a flesh-eating parasitic fly that lays its eggs in the open wounds of warm-blooded animals (often cattle and deer, but occasionally humans).

Unlike most fly maggots, which feed on dead tissue, NWS larva consume and burrow into live tissue, causing substantial morbidity and mortality. (See ASM 2025 Review New World Screwworm: Rise, Fall and Resurgence).

During the 1980s, eradication efforts - primarily utilizing the Sterile Insect Technique (SIT) - expanded NWS elimination southward through Mexico and Central America, to the Darien Gap in  Panama where for many years (starting in 1996) the line was held. 

But in 2023, the number of screwworm detections in Panama surged from a few dozen to thousands, and the infestation began moving northward again. While likely due to a myriad of reasons - including climate change and human migration - supply chain issues during the COVID pandemic is often cited as a factor. 

Since then, the NWS has been moving north steadily, and has been reported getting closer to the U.S./Mexico border in recent months. In August of last year, the HHS reported a Singular Traveler-Associated New World Screwworm Case. 

The USDA has produced its own projections of the economic impact of NWS (in 2024), which finds it could cost Texas farmers nearly 2 billion dollars a year.

Not unexpectedly, last night the USDA announced the first detection of NWS on a farm in Southern Texas. While this is not a big public health concern, should it spread widely, it could prove quite costly to the cattle industry. 

Last night's announcement focuses primarily on reassuring the public (about food safety, and that authorities are responding aggressively). While serious, for now this remains a single detected incursion; not evidence of established transmission within the United States.   

First the announcement from the USDA (h/t to Sharon Sanders on FT for the head's up), after which I'll have a bit more. 

USDA Confirms Presence of New World Screwworm in the United States

Contact: aphispress@usda.gov

Animal Health Officials Working Quickly to Protect U.S. Livestock and Wildlife

WASHINGTON, DC, JUNE 03, 2026 – The U.S. Department of Agriculture’s (USDA) Animal and Plant Health Inspection Service (APHIS) confirmed the detection of a New World screwworm (NWS) in a bovine in Zavala County, Texas. NWS is a serious pest that affects livestock, pets, wildlife, and less commonly, people and birds. NWS larvae (maggots) burrow into the flesh of living animals, causing serious damage to livestock and economic losses.

The affected animal is a 3-week-old calf and larvae were identified in its umbilical area. To date, there have been no further detections.

“All models showed New World Screwworm entering the country in 2025; however, thanks to the hard work across the entire Trump administration and our industry, state, and local partners, we were able to buy time for this moment. Protecting our livestock industry is a national security issue of the utmost importance, and USDA is wasting no time in taking action,” said Dudley Hoskins, Under Secretary for Marketing and Regulatory Programs. “USDA invested heavily in the tools needed to eliminate NWS ever since cases started increasing in Central America and Mexico. The United States has defeated this pest before, and we will do it again.”

USDA and Texas officials are taking immediate action to contain and eradicate NWS from the United States, following the strategies and actions outlined in the NWS Response Playbook. This includes:

• Forming a unified Incident Command Team with the Texas Animal Health Commission and deploying response personnel to the area;
• Establishing a 20 km infested zone around the detection and implementing quarantines, movement controls, and surveillance in this area;
• Expediting targeted release of sterile NWS flies by immediately deploying ground release chambers in the area, in addition to the 4 million sterile flies per week already being released aerially in the area;
• Increasing trapping for NWS flies along the border and just outside of the dispersal area;
• Implementing NWS surveillance and management strategies in wildlife; and
• Conducting targeted outreach in the local area.

Additionally, USDA’s National Veterinary Stockpile stands ready to assist, and will provide resources including treatments, equipment, and logistics support the response as needed.

USDA will continue to work with state departments of agriculture, animal health officials, industry, and producers to mitigate economic impacts of restrictions as much as possible, including negotiating with our trading partners to regionalize any trade restrictions on live animals, limiting them to defined geographic areas.

NWS maggots can infest livestock and other warm-blooded animals, including in rare cases people. They most often enter an animal through an open wound and feed on the animal’s living flesh.

USDA urges residents in the area to check their pets and livestock for signs of NWS. Look for draining or enlarging wounds and signs of discomfort. Also look for screwworm larvae (maggots) and eggs in or around body openings, such as the nose, ears, and genitalia or the navel of newborn animals. If you suspect your animal is infected with screwworm, contact your state animal health official or USDA area veterinarian in charge immediately.

While not common in people, if you notice a suspicious lesion on your body or suspect you may have contracted screwworm, seek immediate medical attention.

The U.S. food supply is safe. Screwworms do not infest meat, fruits, vegetables, or other food sources. USDA’s Food Safety and Inspection Service (FSIS) ensures that the nation’s commercial supply of meat, poultry, and egg products is safe and properly labeled. Under the Federal Meat Inspection Act (FMIA), FSIS inspection personnel must inspect all eligible animal species unless they are exempt or covered by a state inspection program.

Any evidence of screwworm infestation in an animal would be identified during these inspections, and any contaminated product from an affected animal would not be allowed to enter the food supply.

For more than a year, USDA has led a unified response to NWS. As the lead coordinating agency, USDA has deployed advanced surveillance systems and supported robust cross-border response efforts in Mexico and Central America to combat the pest and push NWS away from the United States. These efforts have bought time for USDA to increase domestic preparedness efforts.

Learn more about New World screwworm at Screwworm.gov.

All this comes at a particularly bad time, since U.S. cattle inventories are already at their lowest level in decades (due to prolonged droughts and high feed costs), and beef prices are at record highs. 

 As we've seen previously, the introduction (or reintroduction) of agricultural and human diseases can be difficult to contain. A few examples include:

• the introduction and spread of West Nile Virus in 1999
• the return of Dengue to South Florida after 7 decades
• the post-2000-elimination resurgence of Measles in the U.S. and Canada
• the expansion of Lyme disease across the nation
• and the inexorable spread of Chronic Wasting Disease in North American cervids 

The good news is the USDA has known this day was coming, and has been actively preparing for it. Last April, the USDA released an Updated New World Screwworm Response Playbook, and in January announced the  Completion of Sterile Fly Dispersal Facility in Texas. 

With a little luck, this breach can be contained.  

But as long as the NWS continues to circulate in Mexico, future incursions are all but inevitable. And with all such border interdictions, the defender must succeed 100% time, while the invader only has to succeed once. 

  

Preprint: Increased burden of influenza A/H1N1pdm09 in older adults following the COVID-19 pandemic

#19,187

If there is one constant with influenza A viruses, it is that they are constantly changing; meaning that things that we may have assumed to be true for years or even decades can abruptly change.  

One such `truism' has been that H1N1 viruses pose more of a risk to younger individuals - those born after the 1957 H2N2 pandemic emerged - than to the elderly (see PLoS Path.: Childhood Immune Imprinting to Influenza A).

While older individuals weren't immune to H1N1, they have tended to be less susceptible to infection - and typically saw milder disease - than from the H3N2 subtype.

At least - according to a preprint published last week - until 2022, when a new clade of H1N1 emerged (6B.1A.5a.2a) after the COVID-19 induced lull, which appears to have become more impactful on the elderly. 

Although it was overshadowed by the outbreaks of early May (hantavirus & Ebola), the JID reported a similar finding (see Effects of Age and Birth Cohort on Influenza A Virus Subtype-Specific Hospitalization Rates, United States, 2010–2025) in late April.

 

Assuming these reports are correct, the notion that the elderly might enjoy a bit of a respite during H1N1 centric flu seasons may no longer be true, and that could substantially increase the burden on healthcare facilities. 

An additional caveat: it has not been established whether (or how strongly) this pattern will persist as H1N1 continues to evolve.  Clade 5a.2a has already been largely replaced by newer subclades (see week 20 FluView), so ongoing surveillance and analysis is needed. 

 
First, the link and some excerpts from the preprint, but the full report is very much worth reading.

Increased burden of influenza A/H1N1pdm09 in older adults following the COVID-19 pandemic
Simon P. J. de Jong, Colin A. Russell
doi: https://doi.org/10.64898/2026.05.20.26353664
This article is a preprint and has not been peer-reviewed

Preview PDF

Abstract

Of the two influenza A virus (IAV) subtypes circulating endemically in humans, A/H3N2 and A/H1N1pdm09, A/H3N2 has historically been the dominant driver of disease burden in older adults. Based on an analysis of publicly available global surveillance data from 2015 to 2025 (>300,000 subtyped, age-stratified infections), we report a substantially increased contribution of A/H1N1pdm09 to influenza morbidity in older adults since approximately 2022.

Birth cohort-stratified analyses suggest elevated A/H1N1pdm09 burden among individuals born before 1955-1959, consistent with erosion of pre-existing immunity originally generated by exposure to historical A/H1N1 strains. Pooled estimates across datasets and analytical approaches indicate the increase in A/H1N1pdm09 burden rises with earlier birth year, ranging from 1.22-fold (95% CI 1.08-1.37) for the 1955-1959 birth cohort to 3.10-fold (95% CI 2.58-3.72) for the 1930-1934 cohort. These findings point to a substantial rise in the overall influenza burden among the most vulnerable age groups, with implications for vaccine policy, clinical management, and public health planning.

Main text

Older adults are at markedly elevated risk of severe outcomes from influenza. Of the two influenza A virus (IAV) subtypes currently circulating in humans, A/H3N2 and A/H1N1pdm09, A/H3N2 has historically disproportionately contributed to influenza burden in older adults1,2. However, emerging evidence suggests possible shifts in this epidemiological landscape. A recent US-based study reported increasing contributions of A/H1N1pdm09 virus infections to influenza hospitalizations among older adults since the COVID-19 pandemic3.

        (SNIP)

The apparent increases in A/H1N1pdm09 risk are greatest in the earliest birth cohorts (Fig. 2). This could have substantial clinical and public health implications, as risk of severe influenza disease and mortality increases strongly with age5,6. As such, our results have implications for healthcare capacity planning, with likely substantially elevated acute and post-acute healthcare demand due to influenza compared to a counterfactual in which no immune erosion had occurred. The severity of the 2024/2025 US influenza season potentially reflects the erosion of protection in early birth cohorts3, with adults aged 75 and over experiencing the highest rates of A/H1N1pdm09-associated hospitalization since the 2009 pandemic16.

Vaccines have demonstrated effectiveness against A/H1N1pdm09 among adults aged 65 and over in the post-pandemic period17,18, but new targeted analyses in the earliest birth cohorts, for which estimated increase in burden is greatest (Fig. 2), would be valuable. A potential concern is that individuals whose immunity was shaped predominantly by early-life responses to historical A/H1N1 epitopes may, through repeated boosting of those same responses, lack immunity to epitopes on strains circulating since 2022, leaving them with little cross-reactive protection to recall upon vaccination or infection. These uncertainties notwithstanding, the elevated A/H1N1pdm09 risk suggested here further strengthens the case for vaccination in these birth cohorts and has implications for vaccination strategies, including the value and cost-effectiveness of enhanced formulations which have demonstrated superior immunogenicity and effectiveness in older adults19.

The US and Brazil hospitalization data indicate increased burden of severe disease due to A/H1N1pdm09, but whether this is due to changes in infection risk and/or changes in clinical severity warrants further investigation, particularly given that A/H1N1pdm09 was associated with relatively severe outcomes among hospitalized adults before the COVID-19 pandemic20,21.

Our analyses rely on observational data and are susceptible to reporting biases; however, consistency across four surveillance systems with distinct ascertainment mechanisms and varying analysis methods makes a surveillance artifact implausible. Replication with additional age- and subtype-resolved surveillance data would further substantiate our findings.

Together, our results indicate that the earliest birth cohorts, who are the most clinically vulnerable and have historically been considered relatively protected from influenza A/H1N1pdm09, likely now face substantially elevated risk from these viruses.

This carries potential implications for vaccine policy and healthcare capacity planning. Ensuring that surveillance infrastructure captures age- and subtype-resolved influenza data globally will be essential for tracking this shift and quantifying its consequences for the populations most at risk for severe influenza.

        (Continue . . . )

We tend to think that next year's seasonal flu will be pretty much like last year's, and the year before.  But even non-pandemic flu viruses can throw us curves. 

• In 2008, H1N1 abruptly developed nearly 100% resistance to oseltamivir;

• In 2014, we saw a late summer mutation appear in the H3N2 virus (see HAN Advisory On `Drifted’ H3N2 Seasonal Flu Virus) which greatly reduced the effectiveness of that year's flu vaccine.

• in 2016, we saw warnings over mutated (D225G) H1N1 viruses in Russia and in Europe (see ECDC Risk Assessment : Reports Of Severe A(H1N1)pdm09 In Europe);

• Again in 2025, a drifted (Subclade K) H3N2 emerged which impacted vaccine effectiveness

• And while it has had only limited impact so far, we are now watching another slow rise in antiviral resistance in H1N1.

All reasons why we should give seasonal flu the respect it deserves. 

CMAJ: Acetaminophen in pregnancy

Photo Credit – Wikipedia

#19,186

Last September, following a White House announcement suggesting a link between acetaminophen use during pregnancy and rising rates of autism, we looked at a statement from the World Health Organization and other medical stakeholders which pushed back on the theory. 

While no drug can claim to be 100% safe in 100% of the people that take it, these OTC pain/fever relievers have long been considered the safest option for both the mother and unborn child.

We've looked at potential drivers of increased autism often over the years, and after genetics, environmental exposures - and increased recognition of those on the spectrum - fevers during pregnancy are frequently cited (see Molecular Psy.: Increased Autism Risk Linked To Prenatal Fever).

While a few studies have claimed a statistical link between acetaminophen use during pregnancy and autism (Link), none have found a causal link, and a large 2024 study found `Acetaminophen use during pregnancy was not associated with children’s risk of autism, ADHD, or intellectual disability in sibling control analysis.'

The concern is: disparaging the only `presumed safe' option to reduce maternal fevers could actually end up increasing the incidence of autism, instead of decreasing it.

We revisited this story last January in The Lancet: Prenatal Paracetamol Exposure and Child Neurodevelopment: A Systematic Review and Meta-Analysis, which incorporated 43 studies (17 of which were combined in the meta-analysis), which reported:

Current evidence does not indicate a clinically important increase in the likelihood of autism spectrum disorder, ADHD, or intellectual disability in children of pregnant individuals who use paracetamol as directed, supporting existing recommendations on its safety.

Additional research is still needed to better understand heavy or prolonged usage, but these findings should be reassuring to anyone who chooses to use these drugs during pregnancy.

Under Implications of all the available evidence, the authors wrote:

Taken together with large-scale sibling-controlled studies from Sweden and Japan published in 2024 and 2025, our findings support the safety of paracetamol when used appropriately during pregnancy. They reinforce the guidance of major professional and regulatory bodies, including the American College of Obstetricians and Gynecologists, the Royal College of Obstetricians and Gynaecologists, and the European Medicines Agency, which continue to recommend paracetamol as the first-line analgesic and antipyretic in pregnancy.

Avoiding paracetamol based on inconclusive or biased evidence might increase the risk of maternal fever or untreated pain, both of which can harm pregnancy outcomes. Future research should focus on improving exposure measurement, standardising outcome definitions, and integrating mechanistic and family-based designs to clarify any residual uncertainties.

Admittedly, not a rock solid 100% guarantee of absolute safety; but in life, precious little is. 

Weighing in on all of this, yesterday the CMAJ published a brief review on the use of acetaminophen in pregnancy, and they too find that `Current evidence does not indicate a causal link between acetaminophen use in pregnancy and adverse infant outcomes.'

For those looking for a bit more reassurance, I've posted their report below. 

Acetaminophen in pregnancy
Jonathan S. Zipursky Rachela Smith and Tali Bogler CMAJ
June 01, 2026
DOI: https://doi.org/10.1503/cmaj.260138
PDF

Acetaminophen is the preferred analgesic and antipyretic agent for pregnant people

More than half of pregnant people use acetaminophen, most commonly for pain.1 Compared with alternatives, acetaminophen is the best studied analgesic and antipyretic in pregnancy and has the most reassuring maternal and fetal safety profiles. Major regulatory and obstetrical organizations support judicious use of acetaminophen as the first-line treatment for fever and pain in pregnancy.2

Studies on the risks of neurodevelopmental disorders related to acetaminophen use in pregnancy have not established causation

Some systematic reviews suggest small associations between prenatal acetaminophen use and autism spectrum disorder and attention-deficit/hyperactivity disorder. However, studies that rigorously control for potential confounders or emphasize sibling-controlled designs have typically found weaker or null associations.3,4 This suggests that unmeasured confounders (e.g., genetic and environmental factors) largely explain the associations with neurodevelopmental disorders.5 Randomized controlled trials have not been conducted, and therefore methodological weaknesses in existing observational studies limit conclusions about causation.

Data on fetal reproductive and endocrine effects show small, inconsistent associations

Results from observational studies linking prenatal acetaminophen use to abnormalities in the male reproductive tract are inconsistent; some studies have demonstrated small associations with cryptorchidism and reduced anogenital distance, while others have not.5 Furthermore, most studies relied on maternal memory of acetaminophen use in pregnancy, raising the potential for recall bias that may increase observed effect sizes or lead to spurious associations.

Untreated maternal fever has been associated with adverse fetal outcomes

Separating the potential effects of acetaminophen from those of the underlying condition is challenging. Treatment of fever in pregnancy is clinically indicated; in some observational studies, maternal fever was linked to increased risks of neurodevelopmental disorders.6 First-trimester fevers have been linked to congenital abnormalities such as neural tube, heart, and oral cleft defects.7

Counselling on acetaminophen use in pregnancy should emphasize a critical approach to interpreting the data

Current evidence does not indicate a causal link between acetaminophen use in pregnancy and adverse infant outcomes. Clear counselling by clinicians about the limitations of existing observational data can reduce unnecessary anxiety or avoidance of appropriate fever and pain treatment in pregnancy.

Preprint: Detection of Anti-H5 Antibodies in People with Exposure to Wild Birds in Northern Canada

 

#19,185

Almost exactly 2 years ago - 3 months after the discovery of HPAI H5N1 in U.S. dairy cows - the CDC released the results of their study on Population Immunity to A(H5N1) clade 2.3.3.4b Viruses.  They reported:

CDC analyzed sera (blood) collected from people of all ages in all 10 HHS regions. Blood samples were collected during the 2022-2023 and 2021-2022 flu seasons. These samples were challenged with H5N1 virus to see whether there was an antibody reaction. 

Data from this study suggest that there is extremely low to no population immunity to clade 2.3.4.4b A(H5N1) viruses in the United States.  Antibody levels remained low regardless of whether or not the participants had gotten a seasonal flu vaccination, meaning that seasonal flu vaccination did not produce antibodies to A(H5N1) viruses.

This was in the United States, of course, where exposure to HPAI H5N1 has historically been very low. Over the years we've looked at a number of other seroprevalence studies - particularly in regions with known outbreaks - which have produced slightly different results.

• In 2004 (see The Thailand Serological Study) 322 poultry farmers (in provinces where H5N1 had been detected) were tested. Researchers found that "no poultry workers had microneutralization titers >80, whereas 7 (2%) had lower titers that did not meet the WHO definition for seropositivity".
• In May of 2009 (see Cambodian Study Finds Rare Asymptomatic H5N1 Infections) we saw a study published in the Journal of Infectious Diseases on more than 600 members of a Cambodian village where 2 human H5N1 cases were detected in 2006. Antibody titers showed that only 1% (7 of 674) of the villagers tested had contracted, and fought off, the H5N1 virus. A figure much lower than many had expected.

• In 2012, in H5N1 Seroprevalence Among Jiangsu Province Poultry Workers, we saw a study that found across three locations tested (Gaochun, Jianhu and Gaoyou counties) the percentage of workers testing positive ranged from zero (Gaochun) to 5.38% (95%CI, 2.19%–10.78%) in Gaoyou.

• More recently, in 2024's MMWR: Serologic Evidence of Recent Infection with HPAI A(H5) Virus Among Dairy Workers, in a relatively small sampling (n=115) 7% (n=8)` . . . of exposed dairy farm workers in Michigan and Colorado had serologic evidence of infection with HPAI A(H5).'

Not unexpectedly, people with direct exposure to infected poultry or livestock tend to have higher detectable levels of HPAI H5 antibodies than the general population. There are a few caveats to serological tests, including:

• They only pick up past infections ( >1-3 wks), and are not designed to detect current infections
• Detectable levels of influenza antibodies can wane over time, meaning older or milder exposures might no longer be detectable
• Exposure to some subtypes can sometimes produce cross-neutralizing antibodies to another (see EID Journal: A(H5N1) NA Inhibition Antibodies in Healthy Adults after Exposure to Influenza A(H1N1)pdm09).

In other words - while incredibly useful for retrospective investigations - there has always been some degree of ambiguity regarding the results of seroprevalence testing.

All of which brings us to a preprint, published last week, which looks as a small sampling (n=65) of blood samples collected from people (mostly hunters) with exposure to wild birds in Northern Manitoba, Canada. 

The authors reports a ` 7.4% (n=5/68) anti-H5 seroreactivity rate among hunters in Northern Canada'. Some of these hunters had exposure to small mammals as well.  

The authors list some of the limitations to their study, stating:

The relatively small sample size of the study provides limited power for statistical analyses, and given that the study was conducted in a small region of Manitoba, future larger-scale studies are needed to evaluate how commonly people with exposure to wild birds may be infected by H5N1. 

Unfortunately, the durability of the antibody response to H5N1 is unknown, therefore we can only conclude that the five seroreactive individuals have been exposed to the virus at some during their lifetimes, although the durability of antibodies against avian influenza in human populations is unknown. Whether these individuals experienced symptomatic or asymptomatic infections, and whether they would be protected from subsequent re-infection, is also unknown.

I've reproduced the summary and a few excerpts below, but you'll want to follow the link to read this (not-yet peer-reviewed) report in its entirety.  I'll have a bit more after the break. 

Detection of Anti-H5 Antibodies in People with Exposure to Wild Birds in Northern Canada
Hannah L. Wallace, Morgan Hiebert, Mikayla Hunter, Megan Halbrook, Ryan J. Harrigan, Isaac I. Bogoch, Anne W. Rimoin, Souradet Shaw, Linda Larcombe, Pamela H. Orr, Jason Kindrachuk
doi: https://doi.org/10.64898/2026.05.24.26353994         

Posted May 26, 2026.

        PDF 

Summary

Using a commercially available H5 serology assay, we identified a 7.4% (n=5/68) anti-H5 seroreactivity rate among hunters in Northern Canada. All participants reported close contact with wild birds. 

The Study 

In August 2025 (three months after the 2025 spring goose hunting season), 68 dried blood spot (DBS) samples with matched questionnaires (Appendix 1) were collected from a northern community in Manitoba, Canada. Recruitment was focused on those > 18 years old who were involved in the hunting of birds and/or preparing them (plucking, butchering, food preparation) for consumption.

Some of the participants were also involved in hunting, trapping and handling of mammals (Table 1). Participants were recruited through social media posts and community posters. Participation was voluntary, and included both local community members as well as individuals who were visiting the community for the upcoming fall goose hunting season. Participants were compensated for their involvement in the study. This study was approved by the University of Manitoba Health Research Ethics Board under study HS26941 (H2025-151). 

        (SNIP)

Conclusions

This work indicates that people living and/or hunting in northern Manitoba, Canada have been exposed to H5Nx viruses. Knowledge and concerns about H5N1 among study participants was low, indicating a need for more robust and accessible public health information. Future work including sampling a larger number of individuals and performing longitudinal follow-up to determine how antibody levels change over time would be of significant value. Overall, this work adds to the growing body of literature that aims to understand the true extent of spillovers of avian-origin influenza viruses into humans. 

        (Continue . . . ) 

This isn't the first time we've looked at the potential increased HPAI risk to hunters (and to hunting dogs). Last February we saw Several States Warn On Contact With Wild Birds/Mammals, and the USDA has warned of the risks for several years.

Of course, hunting isn't the only high risk activity.

• In 2024's Mixed Messaging On HPAI Food Safety - we looked at the risks from the slaughtering of live birds and preparation of raw poultry; especially from birds raised at home or purchased from live markets.

• There are scores of LBMs (live bird markets) in the United States (see USDA Report 9 More Live Bird Markets Infected With HPAI H5), and millions of families raise, and slaughter, their own backyard poultry. 

Sadly, the public appears largely unconvinced of the threat from HPAI (see Two Surveys (UK & U.S.) Illustrating The Public's Lack of Concern Over Avian Flu), which almost certainly provides the virus with more opportunities to spread and evolve.