Among patients diagnosed with invasive mold disease at four Atlanta hospitals over five years, 45% died within 90 days, and half required a breathing tube and mechanical ventilation. Those figures come from CDC surveillance published July 30, and they attach hard numbers to an infection that the federal government does not require doctors to report at all.
The report identified 449 cases between 2020 and 2024. Because invasive mold disease is not nationally notifiable, no one has had a reliable local incidence figure for it until now.
"It's kind of the first time we've had an estimate like this," Dr. Jeremy Gold of the CDC, the report's senior author, told the Associated Press.
What the Surveillance Found
Researchers with the CDC and the Georgia Emerging Infections Program reviewed laboratory records from three laboratories serving four Atlanta-area hospitals, two academic, one federal, one community, plus their outpatient clinics. They flagged patients with positive mold cultures or positive Aspergillus galactomannan blood or lung fluid tests.
That produced 968 patients with possible infection. Only 449, or 46%, were ultimately classified as having invasive mold disease. The rest largely reflected colonization or laboratory contamination rather than true infection, which is precisely why this kind of case-by-case review had not been done before.
Cases were classified using established consensus criteria from the European Organization for Research and Treatment of Cancer and the National Institute of Allergy and Infectious Diseases Mycoses Study Group, revised in 2020. Of the 449 cases, 89 met proven criteria, 142 met probable criteria, and 218 were counted as surveillance cases, meaning a treating clinician diagnosed the infection and started mold-active antifungal treatment, or the patient died within three days of testing.
Incidence at the two academic hospitals averaged 4.8 inpatient cases per 100 inpatient beds per year and 14.0 intensive care unit cases per 100 ICU beds. At the community hospital, the figures were 2.8 and 10.2. Aspergillus species accounted for 71% of infections. Two-thirds of cases were in the lungs.
Treatment was widely given. Of the 449 patients, 363, or 81%, received antifungal medication effective against molds, most often isavuconazole, voriconazole, or amphotericin B. The MMWR report notes that high mortality despite high treatment rates suggests new antifungal agents and better treatment protocols could help.
Why This Is Not a Story About Mold in Your House
This distinction matters enough to state plainly. The CDC report opens by noting that mold is common in indoor and outdoor environments and does not usually cause severe infection in people with healthy immune systems.
Invasive mold disease is what happens when mold penetrates tissue in someone whose immune defenses are severely compromised. It is not the same as mold allergy, not the same as asthma triggered by a damp basement, and not what a home inspection finding means for a typical family.
In this surveillance population, 65% of patients had at least one recognized risk factor for the disease. Compared with patients who turned out not to have invasive mold disease, those who did were far more likely to have a blood cancer, 22% versus 3%, prolonged neutropenia, 9% versus under 1%, or a solid organ transplant, 27% versus 7%. Nearly six in ten were taking immunosuppressive medication, compared with 19% of those without the disease.
Notably, chronic obstructive pulmonary disease and asthma were more common among the patients who did not have invasive mold disease. Ordinary lung conditions were not what drove risk here.
There is a legitimate environmental dimension to the broader topic. Dr. Jeffrey Jenks, an infectious disease researcher at Duke University who was not part of the CDC team, told the AP he agrees "the frequency of fungal infections will likely increase with climate change." Gold pointed to a growing population of older adults who are more susceptible. Neither statement means household mold poses this danger to a healthy person.
Who Faces the Real Risk
The highest-risk groups are people receiving chemotherapy for blood cancers, stem cell and solid organ transplant recipients, people on prolonged high-dose corticosteroids or other immunosuppressive drugs, and patients who are critically ill, particularly after severe COVID-19 or influenza.
The COVID-19 finding is one of the report's sharper results. Among the 58 patients with a current or recent COVID-19 diagnosis, 66% were admitted to intensive care in the two weeks before testing, compared with 39% of others, and 90-day mortality was 66% versus 41%. Both differences were statistically significant. The proportion of cases involving COVID-19 peaked at 23% in 2021 and fell to 5% by 2024.
Some risk factors fell outside the classic list. Cirrhosis, end-stage renal disease, and severe burns within the previous 90 days were all significantly more common among patients with the disease. More than a third of cases, 35%, occurred in patients with none of the recognized predisposing host factors at all, which the CDC flagged as the report's central message for clinicians.
Age skewed older. Patients aged 45 to 64 made up 43% of cases and those 65 and older 39%. Children accounted for less than 1%, though the surveillance system did not include a children's hospital.
What the Report Does Not Show
The CDC authors are direct about the limits, and those limits belong near the top rather than buried.
The data come from four facilities in one metropolitan area, which the report says limits generalizability. Two are academic referral hospitals that draw complex patients from long distances, and the authors note this likely pushed mortality higher than in the earlier 2017 to 2019 pilot, which found a 32.7% rate.
The 45% figure is 90-day all-cause mortality among 349 patients, excluding 2024 cases because post-discharge death data were unavailable. All-cause means these patients died of something within 90 days, not necessarily of the mold infection. Most were gravely ill from other conditions.
The surveillance system also missed cases. It did not use DNA-based detection methods during this period, though the report says those were added beginning with 2025 data. It excluded patients with cystic fibrosis, in whom mold colonization is common, and relied on medical chart abstraction. The report presents its incidence figures as benchmark data for future surveillance rather than as a national rate.
This is observational surveillance, not a clinical trial, and it does not establish that any specific exposure caused any specific infection. No treatment guideline changes as a result. Several authors disclosed relationships including research funding and honoraria, and one disclosed a grant from Merck to the CDC.
What Happens Next
The CDC says the findings could serve as baseline data allowing hospitals to distinguish expected sporadic cases from a possible health care-associated cluster. Because incidence is low and severity is high, the report says even small increases should prompt consideration of a common source. Surveillance continues through the Georgia program, one of 12 Emerging Infections Program sites, with DNA-based detection now incorporated.
What remains unknown is whether these rates hold in other regions and in children, and whether earlier diagnosis would improve survival. Invasive mold disease remains outside national notifiable disease reporting, so no comprehensive national count exists.
For readers, the practical guidance is narrow because the risk is narrow. People who are immunocompromised should ask their oncology or transplant team what precautions apply to them, particularly around construction dust, gardening, and water-damaged buildings, and should report new fever, cough, shortness of breath, sinus pain with facial swelling, or an unusual skin lesion promptly rather than waiting. Anyone with a healthy immune system does not need to change anything based on this report, and no one should start an antifungal medication without a clinician.
The bottom line: CDC surveillance found that 45% of patients diagnosed with invasive mold disease at four Atlanta hospitals died within 90 days from any cause, and produced the first local incidence estimate for a disease no one is required to report. People with weakened immune systems and those recovering from critical illness face the real risk. The reasonable action for those patients is a conversation with their specialist about symptoms and precautions. The central uncertainty is whether these rates reflect the country as a whole.
Frequently Asked Questions
What is invasive mold disease? A rare, life-threatening infection in which mold invades body tissue, most often the lungs. It is different from mold allergy or asthma triggered by damp indoor air, and it mainly affects people with severely weakened immune systems.
Does mold in my home put me at risk of this? For a person with a healthy immune system, no. The CDC report notes mold is common in the environment and does not usually cause severe infection in people with competent immune systems. Household mold remains worth remediating for other reasons.
What does the 45% figure actually measure? Ninety-day all-cause mortality among 349 patients with invasive mold disease, excluding 2024 cases. It means those patients died of any cause within 90 days, not necessarily of the infection.
Who is at highest risk? People receiving chemotherapy for blood cancers, transplant recipients, people on prolonged immunosuppressive medication, and patients critically ill after severe COVID-19 or influenza.
Does this apply nationally? Not directly. The data come from four hospitals in one metro area, and the CDC presents them as benchmark figures for future surveillance rather than a national rate.
What symptoms should a high-risk patient watch for? New or worsening fever, cough, shortness of breath, chest pain, sinus pain with facial swelling or a dark nasal lesion, confusion, or an unusual skin lesion. Report these to your specialist promptly rather than waiting.
Should anyone start antifungal medication because of this report? No. Antifungal drugs require a diagnosis and prescription. Do not start, stop, or change any medication without speaking to a qualified clinician.
