Think Differently. Solve the Impossible.

Abstract 

This article explores a question:

What if some people being treated for mold-related illness really are sick—and the treatments really are helping—but mold isn't always the reason they're sick?

Some of the biological changes doctors associate with mold-related illness also appear in conditions such as long COVID and ME/CFS, suggesting that different events—mold exposure, an infection, a toxin, or something else—might leave the body in a similar state of ongoing dysfunction.

The original trigger may be gone, while the symptoms remain. That doesn't mean mold can't make people sick or that these treatments don't work. It means we should separate three questions, as the answers may not always be the same:

What's wrong with the person's body? 
What caused it? 
Why is the treatment helping?

What If We’re Treating the Right Problem for the Wrong Reason?

Introduction

For years, I’ve questioned the use of urinary mycotoxin testing to determine whether someone has been exposed to mold in a home. 

My concern has been fairly straightforward.

Finding mycotoxins in urine does not tell us where those mycotoxins came from. People can be exposed to mycotoxins through food, and urinary mycotoxin measurements are used in scientific research as biomarkers of exposure. The CDC has cautioned that mycotoxins can be detected in the urine of healthy people and that disease-predictive levels have not been established for using these tests to diagnose illness associated with water-damaged buildings.

I've questioned the inference:

Mycotoxins in urine → mold exposure in the home → mold is making the patient sick.

There aresteps between those observations and conclusion.  I've started wondering whether I've been asking the wrong question. What if the urinary mycotoxin result isn't particularly useful for determining whether a home is making someone sick, and some of the physicians treating these patients have nevertheless identified something real?

What if some patients do get better? In at least some cases, what if the treatment works, and for a reason different from the one everyone assumes?I'm not proposing that this is what's happening. I'm asking whether it could be happening.
 

A Biomarker Doesn't Tell Us the Cause

Biomarker. It's easy to give that word more meaning than it has. A biomarker is simply a measurable biological characteristic that provides information about a biological process, condition, exposure, or response.

Blood glucose can be a biomarker.
C-reactive protein can be a biomarker of inflammation.
Blood pressure can function as a biomarker.

A biomarker can tell us something is happening without necessarily telling us why it is happening. That's important when discussing mold-related illness.

A urinary mycotoxin measurement may provide information about exposure to a particular compound. It doesn't establish the source of that exposure. The presence of the compound does not mean it caused the patient's symptoms.
 

Urinary Mycotoxins and CIRS Aren't the Same Approach

I had previously thought of Ritchie Shoemaker's CIRS approach as an approach to treating people for mold exposure. That's not accurate.

Shoemaker does not rely on urinary mycotoxin testing to diagnose CIRS. He has rejected urinary mycotoxin testing as providing conclusive diagnostic information for CIRS. His approach examines what he believes is a characteristic physiological response. The measurements commonly associated with his CIRS framework include:

His framework can also include VCS—visual contrast sensitivity—HLA-DR/DQ genetic susceptibility testing and other measurements. It's important not to call all of these the same.

ERMI/HERTSMI evaluates the environment.

HLA provides genetic information.

C4a, TGF-β1, MMP-9 and the other laboratory measurements evaluate aspects of the patient's physiology.

VCS is a functional test.

Symptoms are clinical observations.
 

Shoemaker Doesn't Say Mold Is the Only Cause

Shoemaker's broader CIRS model isn't Mold → CIRS.

He has proposed that different exposures— water-damaged buildings, ciguatera and other biotoxin-related exposures—can initiate a similar persistent inflammatory condition. His research includes people whose illnesses were attributed to different initiating exposures.

An interesting example is ciguatera, an illness caused by eating fish contaminated with ciguatoxins. Shoemaker and colleagues conducted a case-control study involving 59 patients with chronic illness following ciguatera and 59 controls.

They measured VCS, HLA, VIP, MSH, C4a, TGF-β1 and other inflammatory parameters and reported abnormalities overlapping with some they had identified in other populations they classified as having biotoxin-associated illness.

That raises an interesting possibility: Different initiating events might produce some similar downstream physiological abnormalities. It doesn't establish that biotoxins are the thing connecting all of them.
 

What Happens When We Look Outside CIRS?

Let's take the measurements Shoemaker associates with CIRS and look for them in diseases that aren't defined by mold or water-damaged-building exposure. Several of these abnormalities appear elsewhere:

C4a

C4a is produced during activation of the complement system, part of the body's immune response. Shoemaker associates elevated C4a with CIRS. But complement activation, including C4a abnormalities, has also been reported in ME/CFS and long COVID. That means elevated C4a doesn't tell us what activated the system.

TGF-β1

Shoemaker associates elevated TGF-β1 with CIRS. TGF-β1 is involved in immune regulation, inflammation, tissue repair and fibrosis. Abnormalities have also been reported in ME/CFS and numerous other inflammatory and disease processes. The measurement may be real. What does it mean?

MMP-9

Shoemaker associates elevated MMP-9 with CIRS. MMP-9 participates in inflammation, extracellular-matrix remodeling, vascular biology and blood-brain-barrier processes. Elevated MMP-9 has also been reported independently in long COVID.

VEGF

Shoemaker's model has included abnormalities involving VEGF, particularly low VEGF in certain CIRS patients.Reduced circulating VEGF-A has also been reported in ME/CFS. VEGF biology is complicated. Different studies measure different forms under different conditions. Again, we're seeing an overlap in the physiological system.

ACTH and Cortisol

Shoemaker examines ACTH/cortisol relationships as part of his CIRS framework. HPA-axis abnormalities have been studied extensively in ME/CFS, and cortisol abnormalities have also been reported in long COVID. A remote analysis of ME/CFS research found evidence consistent with a hyporeactive HPA-axis state, including lower morning salivary cortisol, lower 24-hour urinary cortisol and impaired responses under certain stimulation conditions.

MSH

This one is interesting as it doesn't neatly support the overlap hypothesis.Shoemaker commonly associates CIRS with low alpha-MSH. One independent ME/CFS study found higher, rather than lower, plasma alpha-MSH, particularly earlier in the illness. If different diseases produce biomarkers moving in different directions, that could help distinguish them.

VIP and ADH/Osmolality

Shoemaker also uses VIP and the relationship between ADH and osmolality. There is research into related neuroendocrine, fluid-regulation and autonomic abnormalities in other chronic illnesses. I haven't found convincing evidence that ME/CFS or long COVID reproduces Shoemaker's proposed VIP and ADH/osmolality patterns. Those remain questions, rather than established overlaps.
 

Individual Biomarkers Aren't the Same as a Biomarker Pattern

There's a limitation to what I've described.

Showing that C4a occurs in another disease doesn't disprove CIRS. Neither does finding MMP-9 abnormalities in long COVID or cortisol abnormalities in ME/CFS. A diagnostic pattern can be useful, even when its individual components aren't specific.

Imagine four measurements: A, B, C and D. Each could occur independently in dozens of diseases. Perhaps, A high + B low + C high + D low, occurring together, identifies a particular physiological state. That's possible.

The question isn't, do other illnesses have abnormal C4a. The question is: Does Shoemaker's complete pattern distinguish patients he identifies as having CIRS from patients with other chronic illnesses? I haven't found the study I'd like to see answering that question. Take people with:

• CIRS associated with water-damaged buildings
• long COVID
• ME/CFS
• other chronic inflammatory illnesses
• healthy controls.

Run the same panel under the same laboratory conditions and determine whether the CIRS pattern reliably distinguishes one group from another. Without that comparison, there's a difference between saying: “CIRS patients frequently have these abnormalities” and “this pattern identifies a distinct biotoxin-driven condition.” The first does not establish the second.
 

Long COVID Gives Us an Interesting Way to Think About This

Long COVID provides a useful analogy as we often know approximately when the illness began. A person is healthy. They contract SARS-CoV-2. They become ill. The acute infection resolves. They never return to their previous state of health.

Whatever is producing the continuing symptoms doesn't require the original acute infection to still be occurring. Conceptually:

Trigger -> physiological disruption -> failure to return to baseline - > persistent symptoms

What if something similar happens following other insults? Maybe for one person the initiating event was exposure associated with a water-damaged building; For another, it was an infection. For another, a toxin. For another, something different. Years later it becomes difficult or impossible to determine what started the process.
 

The Cause and the Condition May Not Be the Same

There are two questions: 

What started the illness?
What is keeping the person sick now?

Those don't necessarily have the same answer. A possible model might look like this:

Trigger X -> immune/inflammatory disruption -> autonomic / vascular / neuroendocrine / metabolic abnormalities -> persistent symptoms.

The trigger could disappear while the downstream physiological condition persists. If so, constantly searching for the original trigger might be less useful than determining what remains abnormal in the patient.
 

Where Environmental Mold Testing Fits

This doesn't make environmental investigation irrelevant. If someone is continuing to experience an environmental exposure that is making them sick or preventing recovery, identifying and correcting that exposure is important. Environmental testing answers an environmental question: Does this building have a mold or moisture problem?

Patient testing answers a different question: What is happening physiologically in this person?

There is a third question: Did the first cause the second?

Those shouldn't be collapsed into one conclusion. A building can have a mold problem. A person can have abnormal inflammatory biomarkers. Having both simultaneously does not establish that the building caused those abnormalities. Chronology, exposure history, alternative causes, response to changes in exposure and other evidence matter.
 

ERMI and HERTSMI

Shoemaker uses environmental testing as his model considers water-damaged buildings one potential initiating or perpetuating exposure.

An ERMI or HERTSMI result isn't measuring the patient's physiology. An abnormal physiological marker isn't measuring the building. Connecting those two requires a causal inference. This is why it's important for patients to understand exactly what question each test is capable of answering.
 

Urinary Mycotoxins

The presence of a mycotoxin or metabolite in urine can indicate exposure. The result can be influenced by factors including the particular mycotoxin, diet, timing, metabolism, urine concentration and excretion.

Different mycotoxins have different toxicokinetics. Some dietary mycotoxins can be absorbed and eliminated quickly. For example, controlled human studies have used urinary biomarkers specifically to measure dietary exposure to deoxynivalenol (DON).

A urine result raises questions:

What compound was detected?
What are its common exposure sources?
What foods has the patient eaten?
How quickly is this compound normally metabolized and excreted?
Was this a spot urine sample or a 24-hour collection?
What level has been demonstrated to predict disease?
What evidence connects this result to the patient's home?

Those are not objections to measurement. They're questions about interpretation.
 

What If the Treatment Works?

This brings me back to the question that started this investigation. Suppose a physician treats a patient believed to have mold-related illness. The patient gets better. That's important. We shouldn't dismiss the patient's improvement simply because we question the proposed mechanism.

Improvement leads to the question: What made the patient better?

Treatment programs do more than remove someone from a moldy environment. Depending on the practitioner and protocol, interventions may affect inflammation, immune activity, gastrointestinal function, bile metabolism, nutrition, sleep, vascular function, neuroendocrine regulation or other physiological processes.

Suppose one or several of those interventions correct something that was keeping the patient sick. The treatment could work. The fact that the treatment worked wouldn't necessarily establish what caused the illness.

That gives us something to consider:

What if some clinicians have discovered treatments that genuinely help certain chronically ill patients, and the treatments work by correcting downstream physiological abnormalities rather than by treating the particular environmental cause attributed to the illness?
 

It Could Explain Difficult Cases

Consider someone who has symptoms attributed to mold exposure and repeatedly fails to find a mold problem in their home. Perhaps they've tested multiple environments, and there's no history of a major water-damaged-building exposure. And they're still sick. 

One explanation is that an exposure hasn't been found yet.  Another possibility is: There isn't an ongoing environmental exposure. Perhaps there was one previously, or perhaps something else initiated the illness entirely.

If the model begins with: “You're sick because you're still being exposed,” then failure to recover can lead to an endless search for the exposure. The house gets tested again. Minor findings are remediated. Possessions are discarded. The patient moves. The next house is tested.

If the person remains sick, the conclusion may become: There must still be mold somewhere. That's a hypothesis that risks becoming difficult to falsify. Sometimes continued investigation may indeed uncover a missed environmental problem. The prudent questions to consider are:

What if the thing that started the illness is no longer the thing keeping the patient sick?
What if mold wasn't what started it?

What We Don't Know

I am not concluding:

This investigation does not establish that CIRS isn't real.
It doesn't establish that water-damaged buildings cannot produce chronic illness.
It doesn't establish that CIRS, ME/CFS and long COVID are the same.
It doesn't establish that Shoemaker's biomarker pattern is nonspecific.
It doesn't establish that physicians treating patients for mold illness are unknowingly treating some other disease.

I am interested in the following :

Shoemaker and others have identified measurable physiological abnormalities in some chronically ill patients. Some of those abnormalities have independently been found in chronic illnesses not defined by mold exposure. What's uncertain is what those abnormalities tell us about cause.
 

Questions I'd Ask If I Were Being Treated for Mold Exposure

If you're seeing a physician who believes mold or a water-damaged building is contributing to your illness, I wouldn't use this article as a reason to stop treatment. Use it as a reason to ask better questions:

What evidence indicates that I have an abnormal physiological condition?

Which findings are specific to the diagnosis you're giving me, and which occur in other illnesses?

What evidence indicates that mold or a water-damaged building caused those abnormalities?

Could another infection, illness or physiological event produce the same findings?

If you're using urinary mycotoxins, what does the result establish about the source of exposure?

How have dietary sources been considered?

If you're recommending ERMI or HERTSMI testing, what exactly will that result tell us about my health?

If my home doesn't have a significant mold problem, what other causes will we investigate?

What part of the treatment is intended to remove an exposure, and what part is intended to correct my current physiology?

Most important:

If I improve, how will we determine why I improved?
 

Maybe We're Asking Different Questions

I've spent time investigating buildings, trying to answer questions such as:

Is there a moisture problem?
Is mold growing somewhere it shouldn't be?
Where is it?
What needs to be corrected?

Those are environmental questions.

A physician has a different problem: Why is this person sick, and what can we do to make them better?

These types of investigations overlap. One of the mistakes is trying to make one answer the other. A mold problem in a building doesn't automatically explain a person's illness. An abnormal biomarker in a patient doesn't automatically tell us what, if anything, is wrong with their building.  A treatment that makes someone better doesn't automatically tell us what made them sick.

After doing research on the topic of treatments for chronically ill patients and physiological abnormalities, I'm left with a question, rather than a conclusion:

What if some of the physiological abnormalities being attributed to mold are real, some of the treatments do help, but in at least some patients we've identified the wrong reason they're sick?

That possibility doesn't diminish the patient's illness or a treatment that helps them. It means we need to keep asking a question that matters in both medicine and environmental investigations:

What does the evidence actually establish, and what are we assuming from it?

About the Author

Daniel Stih (danielstih.com) is an aerospace engineer, software engineer, indoor environmental consultant, and author of 12 books. For more than 30 years, he has investigated complex problems spanning engineering, technology, the built environment, and human decision-making. His work explores how evidence, assumptions, and systems shape the conclusions we draw, and whether we're solving the right problem. Learn more about his approach in, Why I Think This Way.