Indoor Fungal Concentrations

1) The commonly used method of comparing indoor and outdoor spore concentrations is invalid.

2)It is impossible to determine the indoor airborne spore concentrations using the spore trap methods typically employed.

3)Virtually none of the decisions based on airborne spore concentrations results are valid.

This discussion addresses the problems associated with the rationale and sampling protocols used by the majority of individuals who attempt to perform air monitoring for moulds in buildings. A discussion on the myths of “clearance sampling” can be found here (“clearance sampling” is the junk-science “testing” practice employed by poorly trained individuals during mould remediation projects.)

Data Quality Objectives
Unquestionably, the greatest flaw in the overwhelming vast majority of air monitoring that is performed is a lack of Data Quality Objectives (DQOs). DQOs are a priori conditions and decisions made before the sampling begins. The methods used for the sampling must meet the necessary detection limits, and must be capable of answering the question being posed.

All too often, we encounter those who incorrectly think the reason they are collecting air samples is to answer the question: “Is there mould in the air?” Since the answer to that question is always “Yes,” then why waste the money on the samples?

Air sampling is NOT a part of a typical initial mould investigation. Air sampling for moulds in buildings is virtually never needed on initial investigations. If a consultant begins collecting air samples in your building or home as one of their first activities, that is a good sign that they are spending your money without having clearly defined objectives in their investigation.

Prior to beginning an air sampling scheme, the most important step is properly defining the question that is really being asked and establishing DQOs that will answer the question. The DQOs are essentially the quantified answer to the question. Armed with the DQOs, one can then select the most appropriate sampling and analysis procedure; keeping costs to a minimum, while still achieving the DQOs.

Typically, the DQO is not to answer the question: “Am I exposed to mould?” (To which the answer is always “Yes,”) but rather, to answer the question: "Am I exposed to a concentration greater than________?" Where the blank will be filled in by the consultant prior to sampling. And typically, the blank will be: "...the corporate safety policy limit...", or "...a level considered dangerous by my physician..." or “…the normal expected concentration for non-symptomatic houses in my area” or similar. The person collecting the sample should then be able to give the numerical value, up front, that they will be using to justify the DQO and they should be able to give you, very clearly, the rationale behind the justification. (For an example of the DQO process, you can see the DQOs that we prepared for the State of Colorado in Colorado’s Assessment Sampling and Analysis Protocol for clandestine methamphetamine labs; open the document and scroll down to Attachment to Appendix A).

Building Conditions
In general, the building should be in a condition that emulates the normal occupied conditions of the building. There is no point in sampling building conditions that don’t normally exist when the building is occupied (unless the DQOs address the non-standard building conditions).

For purposes of classifying conditions during microbial sampling, structures are considered to be in one of two modes: “open building conditions” or “closed building conditions.” In “open building conditions” the influx of outdoor air has a significant influence on the indoor air. In “closed building conditions” the building is sufficiently isolated with conditioned air such that the outdoor air does not significantly influence the indoor air. Most commercial buildings operate under closed-building conditions, and most homes are maintained under closed building conditions.

The mode of the sample building must be known prior to sampling and quantified during sampling. Without this very important information, the very best air sampling methods and results may be utterly useless.

Unless all the doors and windows are open and there is a breeze blowing through your building (literally), ask your consultant how they intend to quantify outdoor influences on the indoor air. If they can’t answer the question, you shouldn’t let them spend your money.


Indoor versus Outdoor Concentrations
Much of the literature concerning the quantification of indoor airborne fungi recommends that indoor samples should always be accompanied by outdoor samples. These recommendations are based on the lack of consensus standards for airborne fungal counts (i.e., no direct comparison to “acceptable” levels can be made), and the unfounded (indeed erroneous) premise that indoor spore concentrations should always be lower than outdoor concentrations and the erroneous notion that there is good correlation between strength of outdoor contaminants and indoor concentrations.

For a start, too many broad statements are made without due considerations for regional and microclimate changes that can greatly alter the variations in concentrations, and can greatly alter the relationship between the indoor environment and the outdoor environment.

For example, consider fungal data collected from the Rocky Mountain region at latitudes spanning from about 35 degrees north to 45 degrees north versus samples collected from Bucerias, Mexico, located on the same latitude as the Hawaiian Islands, with its sub-tropical climate. For the most part, populations in the northern climes have significant seasonal changes and their practices of indoor environmental control changes with those seasons. Whereas in Bucerias, many buildings have no windows whatever, and doors are merely gates; the indoor to outdoor relationships are vastly different.

As such, unless one already has defined the indoor to outdoor relationship for a particular location, for a particular time of year under specific weather conditions, under specific building conditions, how can one compare an indoor to an outdoor count?

Our data indicate that for samples collected under “closed building conditions,” regardless of the region, there is poor correlation ¹ between indoor and outdoor fungal profiles (genera, species and total counts) and regardless of whether the building is a “problem” (symptomatic) building or a “healthy” (non-symptomatic) building. Studies by other researchers ² have made similar conclusions regarding outdoor versus indoor influences, particularly with regard to particulates.^1a

We will provide data on this issue in a section below, in the meantime, in the graphic below, we have presented data for indoor and outdoor fungal concentrations for both symptomatic (problem) buildings and non-symptomatic (healthy) buildings.

Actual statistics aside, as can be seen from the graphic, there are clearly times when the outdoor concentrations of spores are less than the indoor concentrations for healthy buildings, and conversely, there are clearly times when the spore counts indoors are less than outdoors for contaminated buildings.

Viewed another way, ranking outdoor spore counts (graphic below), we see that if we used the outdoor spore count as our exclusive "standard" criteria, in many cases (those on the left of the crossover) we would have "mouldy" houses regardless of how clean or "normal" the spore count, and in many cases (those on the right of the crossover) we would have to classify even the mouldiest of houses as "clean."



Literature containing studies suggesting that indoor and outdoor comparisons should be made do not hold up to closer inspection. For example, occasionally the Shelton study is cited, ³ wherein the authors evaluated the results of several thousands of indoor fungi results and several thousands of outdoor results. And concluded that investigators should collect outdoor samples for comparison. However, the study provides no support for the conclusion. The study itself is a metaanalysis of the work of others, over which, the Shelton team had no control or even knowledge. The premise in the paper was that the samples were collected using standardized methodologies – however, the presumption is unsupported and standardized methods were not used – only presumed. Additionally, of the thousands of anonymous samples that had been collected by other people using unknown methods under unknown circumstances, only 82 of those clients were contacted and given a “brief telephone interview” to learn more about the samples which they had collected.

The net result is that there were thousands of outdoor samples collected (by someone else) and thousands of indoor samples collected (by someone else), and a correlation was attempted. But nowhere in the study did the authors actually study the indoor results vis-à-vis the outdoor results for any specific building. Rather, the large numbers of samples were pooled into arbitrary seasonal and geographical compartments, and the trends over those large pooled data were compared. Temporal differences in the data alone could have been enormous. For example, the unknown consultant on a specific building investigation could have collected a single indoor sample in the building on a sunny Monday afternoon (outdoor temperatures balmy 80F; windows open), and collected the outdoor sample on the following cloudy Wednesday morning (chilly 40F with heavy rains and wind, doors and windows closed and the heating system running) and since the samples were received from the same building, the data would have been pooled together. Hardly a good comparison. So, if anything at all, the Shelton data represent a good source of information about outdoor fungal counts, within certain limitations, but certainly do not speak to indoor counts and certainly do not provide the foundation for a comparison in an individual building.

Outdoor air spore concentrations are not an independent variable. Basing the quality of the indoor aerobiological environment on outdoor comparisons, results in a moving target that is the result of very complex factors that are independent of the indoor air. Using outdoor counts, in a vacuum, as a comparison point simply lacks validity, and is mostly used by consultants who don't understand aerobiology, and who shouldn't be taking air samples.

Fungal Air Sampling Methods
There are no completely accepted protocols for the analysis of biota in air. Different situations will call for different protocols. And different methods will bias certain genera of fungi high or low depending on the particle size of the spores. (Some excellent work on the subject of particle size collection profiles has been performed and reported in searchable peer reviewed literature by Tiina Reponen, PhD, Department of Environmental Health, University of Cincinnati.) Direct comparison of results from one sampling method to those results of another sampling method is not possible. Unless the sampling methods and analysis methods are identical, one cannot directly compare data.

Based on our experience, the two most common methods used while sampling aerobiology, are the Anderson method (for viable organisms) and spore trap methods such as the Air-O-Cell method (for total airborne fungi.)

In our opinion, viable counts carry far less weight than do total counts (such as the Air-O-Cell method). For a start, the sampling error associated with aerobiology is already huge but there are additional errors associated with the Anderson collection technique. Therefore, even comparing Anderson results with Anderson results may be difficult and misleading. In any event, since the offending epitope (allergen) will be associated with the spore, regardless of whether that spore is dead or alive, performing viable sampling (samples reported as CFUs –colony forming units) will always bias the spore count low, and underestimate exposure.


Distribution of Aerobiology
Regardless of the method selected, total counts of airborne fungi exhibit a spatial and temporal lognormal distribution throughout the day.⁴ This means that the variation about the mean can be very high and skewed in one direction. A consultant who collects one or two or even three air samples from a building is doing little more than guessing at the average concentration within the building. See the graphic below.

To properly perform air monitoring, one needs to collect a sufficient number of samples to allow them to determine the statistical distribution and from the distribution curve calculate the point on the probability curve that best represents the “average” concentration in the home or building. For our air monitoring, we typically calculate the minimum variance unbiased estimate (MVUE) which is a point along a lognormal distribution curve that best approximates the “average” of the curve. (One could similarly log-transform the lognormal data and if a good fit Gaussian distribution emerges, one not only confirms the lognormality of the original data, but one can then calculate the arithmetic mean from the log-transformed data. Although the mean of the log-transformed data will be biased low, sometimes extremely low, the method is easier to calculate than the MVUE.)

For example, in one “healthy” building we looked at, the MVUE fungal count was 218 counts/m3 and the data exhibited a classic lognormal distribution (the figure above is the probability curve for the data).

With an MVUE of only 218 counts/m3, we can be very confident the building does not have a fungal problem. However, given the nature of the statistical distribution, there is a 20% probability (one sample in five) that a random sample collected from this building will have a fungal count in excess of 1,000 counts/m3. There is a 10% probability that a single sample will exceed 2,500 counts/m3. If a “cook-book” consultant unfamiliar with microbial populations and statistical analysis enters the building and collects one air sample, they have a very high probability of misclassifying the building as a problem building and unnecessarily alarming the occupants and inappropriately disrupting the lives of those people.

The flip side is failing to identify a building with a serious problem. In one very sick building we investigated, the MVUE count was 9,500 counts/m3. Yet, one of the air samples collected from this apartment was only 316 counts/m3. If an unassuming consultant collected a sample during this time-frame, they could have overlooked what was a very serious problem.

Species Identification
Some cook-book consultants attempt to make species identification rankings between complaint and non-complaint areas and/or indoor and outdoors. In general, the interpretation of species ranking is hit and miss, and largely guesswork. For a start, others have reported that approximately 50% of fungal species identification reported by laboratories using optical identification methods are incorrect. ⁵ Furthermore, it has been observed by qPCR analysis that culturable samples may underestimate some species by three orders of magnitude. Secondly, spatial variations alone can result in widely different results in the same structure.

In two studies^6,7 that looked at actual species identification in house dust, the authors used qPCR for species identification. Data were collected from different cohorts of buildings for the studies. What the authors found was that there was no significant differences in the concentrations of any individual species of moulds between mouldy houses and reference houses. In general, for species ID comparisons to be made, the data must meet “PARCC” parameters (the precision, accuracy, representativeness, comparability, and completeness) of very narrowly defined DQOs. In order for this to happen, the study would be very involved, very costly, and not likely to occur in the vast majority of fungal investigations. This is not to say that species comparisons cannot be made, but they should only be performed within the limitations of the study underway.


What Are “Normal” Spore Counts?
Based on our experience, the normal airborne concentration varies significantly with the geographical latitude of the study building; “Normal” indoor counts in Bucerias, Mexico will be significantly higher (by orders of magnitude) than “normal” indoor counts in Boston, Massachusetts.

As a general rule, the normal closed-mode indoor total fungal spore counts across the central portion of the U.S. (bounded by a latitude of, say, 35 degrees north to 45 degrees north), for healthy buildings (buildings not experiencing fungal problems) is usually less than 500 counts per cubic meter (counts/m3); with indoor concentrations exceeding 900 counts/m3 less than 15% of the time.⁸

Based on our experience (for the same geographical boundaries) closed-mode indoor fungal counts in buildings whose occupants are experiencing symptoms attributed to fungal problems and/or for buildings which have widespread mould growths in the crawlspace or living areas, the MVUE is 2,280 counts/m3 and fungal concentrations exceed 900 counts/m3 greater than 60% of the time. The fungal concentration for symptomatic buildings we have visited in Colorado have been in excess of 5,000 counts (in California, we have observed fungal fragment concentrations of 1,700,000 counts/m3 in symptomatic buildings).

According to our database, the viable fungi concentrations of non-symptomatic, healthy environments is not dissimilar; 383 colony forming units per cubic meter of air (CFU/m3), with a 95% probability that one sample in five (21%) samples will exceed 900 CFU/m3.

Let’s revisit the indoor vs. outdoor question again. The outdoor spore concentrations change dramatically with season, and exhibit a bi-modal distribution whose peaks are in the early spring and early autumn. Unfortunately, our definition of season (summer, autumn, etc.) is based upon the equinoxes, but moulds are more interested in mean diurnal temperatures and precipitation levels, and so there is more correlation between mean diurnal temperatures than lunar seasons. Nevertheless, our database indicates that (for the previously mentioned latitudes) the total fungal indoor counts in non-symptomatic closed-buildings is not influenced by the outdoor fungal concentrations.

In the data set provided (latitude and altitude defined) Winter outdoor MVUE fungal counts average about 209 counts/m3, with individual counts exceeding 900 counts/m3 approximately 10% of the time. Spring counts are approximately 900 counts/m3 with individual counts exceeding 900 counts/m3 approximately 40% of the time. The summer counts average about 3,500 counts/m3 with individual counts exceeding 900 counts/m3 approximately 72% of the time. And yet, non-symptomatic indoor counts remain roughly the same throughout the year (horizontal central line).

Therefore, in normal healthy houses in the latitudes identified, the indoor count will exceed the outdoor counts almost four months out of the year. Laboratories and consultants who claim that a problem exists if the indoor counts exceed outdoor counts ignore the fact that the spore counts in healthy houses stays relatively stable but the outdoor seasonal counts go up and down! Therefore, according to some consultant’s reports we have reviewed, healthy houses contain excessive levels of moulds in the winter but are normal in the summer, even though the spore counts in the houses have not changed.

Not only is there no valid indoor/outdoor relationship (out side the context of specific DQOs), the outdoor variability is similar in context with the indoor variability. Consider the following data which were collected form an actual case:

Air Monitoring Data

Time	Indoor Spore Count	Outdoor Spore Count
10:00	971	6
13:15	16	112
15:23	33	102
18:06	426	133

Using the exclusive criteria that indoor counts should be less than outdoor counts, does this house have a mould problem or not? What if four “certified” mould inspectors were independently hired and visited the house at each of the times listed – what would each conclude?

(Certified mould inspectors are not actually certified in anything, and usually have virtually no legitimate training or knowledge in mycology, aerobiology or microbiology. The “certification” process is not recognized, and certificates are frequently nothing more than what the inspector has printed off their own computer. As a general rule, “certified mold inspectors” represent the lowest quality inspection and assessment services, and usually lack scientific foundation, in favor of the more lucrative “toxic” mould agenda.)

For a discussion of mould and its occurrence in properties complete with many photographs, visit our“Habits” page.

Wrap-up
Sampling, analysis and interpretation of aerobiology requires considerable knowledge in microbiology, sampling theory and statistical applications; it is not cook-book, black-box stuff and should not be performed by home inspectors (or industrial hygienists) who lack the necessary knowledge to understand the limitations.

All kinds of fungal and Bacteriological biomarker analyses including LPS layer analysis (endotoxins) for the Gram negatives, PBHAs, phospholipid ester-linked fatty acids, mycotoxins, volatile organic compounds (VOCs also known as MVOCs and UMVOCs when derived from fungi) and polymerase chain reaction amplification is used in microbial assessments.

In conclusion, consultants who are attempting to characterize airborne microbials based on just a few samples or “indoor v. outdoor” comparisons can't reliably produce meaningful results or data. Do-it-yourself mold testing kits, such as the settling plates marketed by Pro-Labs^® and sold through Home Depot^® and other retail outlets, are entirely and completely incapable of producing any legitimate results under all circumstances and cannot be interpreted by anyone under any known conditions.

References

1a Preventing Occupational Respiratory Disease from Exposures caused by Dampness in Office Buildings, Schools, and Other Nonindustrial Buildings, DHHS (NIOSH) Publication No. 2013–102, November 2012

1 Least square fit, R2 = 0.058

1a El-Hougeiri N., El-Fadel M.IAQ Characterization In Urban Areas: Indoor To Outdoor Correlation Proceedings: Indoor Air 2002

2 Cooley J.D.; Wong W.C.; Straus D.C.; Jumper C.A. Correlation between the prevalence of certain fungi and sick building syndrome, Occupational and Environmental Medicine, September 1998, vol. 55, no. 9, pp. 579-584(6)

3 Shelton BG, Kirkland KH, Flanders WD, Morris GK, Profiles of American Fungi in Buildings and Outdoor Environments in the United States Applied and Env. Microbiology April 2000 pp 1743-1753

4 Shapiro-Wilk W test greater than 0.9913

5 Dean TR, Betancourt D, Menetrez M; Identification of Fungal Species by Multiplex PCR Assay Lecture given at the ASTM International Symposium “Mold in the Indoor Environment: Assessment, Health and Physical Effects, and Remediation” University of Colorado, Boulder, Colorado, August, 2004

6 Vesper SJ, Varma M, Wymer LJ, Dearborn DG et al Quantitative Polymerase Chain Reaction Analysis of Fungi in Dust From Homes of Infants Who Developed Idiopathic Pulmonary Hemorrhaging Journal of Occ. Environ. Medicine Vol. 46 No. 6, June 2004, p. 596-601

7 Meklin T, Haugland RA, Reponen T, Varma M, et al, Quantitative PCR Analysis of House Dust Can Reveal Abnormal Mold Conditions J. Environ. Monit., 2004, 6 (7), 615 – 620

8The last time we analyzed the data, our data shows the MVUE to be 394 counts/m3 with a 13% probability of a single random sample exceeding 900 counts/m3. n=245

This page is a "living discussion;" originally created on October 12, 2002, it was most recently revised on January 26, 2012.

---

To return to our state-of-knowledge mould page, click here.

Visitors to this page generally have an interest in scientific issues. If you are interested in such matters, you may find some of our other discussions interesting.

A discussion concerning myths surrounding duct cleaning, can be found by clicking here.

For a discussion concerning indoor air quality, click here.

For issues surrounding the history and cause of carpal tunnel syndrome click here.

For a discussion concerning indoor radon click here.

For a discussion concerning the myths associated with laboratory fume hood face velocities click here.

For a discussion concerning laboratory fume hood evaluations, click here.

Finally, for a listing of documents associated with the State of Colorado regulations concerning methamphetamine laboratories (meth-labs), click here.

Visit our main page!

Forensic Applications Consulting Technologies, Inc.

Feel free to send Forensic Applications, Inc. an email directly by clicking here