Is ERMI Testing Being Used For Its Intended Purpose? by Gary Rosen, Ph.D.

Is ERMI Testing Being Used For Its Intended Purpose? by Gary Rosen, Ph.D. www.mold-toxins.com 2/28/2015

Copyright 2015 Gary Rosen, Ph.D. & Certified Mold & Allergen Free, Corp. www.mold-free.org This article may be reproduced or transmitted in its entirety with copyright notice by any means, electronic, mechanical, including photocopying, recording, or by any information storage or retrieval system. www.mold-toxins.com 2/28/2015 Page 2

We show in this report that ERMI testing is not being used for the purpose it was developed leading to false conclusions. ERMI was developed as a tool for analyzing a home s history of water damage by measuring the level of mold spores and mold fragments in (old) carpet dust for the purpose of determining if early childhood mold exposure could impact a child s health. In this regard ERMI has been (or would appear to be) successful. Numerous research studies have shown that high levels of mold in carpets correlates with an increase in childhood asthma. However, home owners, often based on doctor recommendation, are now using ERMI testing to determine current conditions in a home or office rather than using it for its original intent which was to analyze a residence s history of water damage. Evaluating current conditions is not the intended purpose of ERMI nor will applying a technique developed to study the history of water damage in a home be reliable for making current assessments. As we shall see, the research published by the developers of ERMI quite clearly show that ERMI analysis of carpet dust (or floor dust) does not correlate with current conditions. We will review this research and explain why ERMI (or HERTSMI-2 for that matter) should not be a consumer tool to evaluate current conditions in a home or office. Others have come to similar conclusions. See: Michel Pinto at: www.wondermakers.com/portals/0/docs/ermi%20for%20post-remediation.is.pdf Stephen Parkhurst at: www.experts.com/articles/mold-dna-detection-can-ermi-be-improved-by-steven- Parkhurst Danielle Dobbs at: http://www.molddetectionexperts.com/newsletter0601.htm Caoimhín P. Connell at: http://forensic-applications.com/moulds/ermi_thread.pdf www.mold-toxins.com 2/28/2015 Page 3

ERMI History Environmental Relative Moldiness Index 1 (ERMI) is a relatively new method of mold analysis invented/ developed by the US EPA and available through a number of commercial labs. ERMI is based on DNA analysis of mold extracted from carpet dust. Traditional mold sampling, still widely used today, is based on culture methods or visual ID of spore characteristics under a microscope. A good description of the traditional mold sampling methods along with pros and cons can be found here. 2 Traditional mold sampling is based on taking indoor air samples and an outdoor air sample and comparing indoor levels of mold spores to outdoor spore levels. If the indoor air spore count is elevated compared to the outdoor and/or the mix of mold species is different between indoor vs outdoor the home would be considered to have elevated mold levels. However when comparing to outside air, the classification of an individual home as moldy or not is complicated by the correlation of indoor levels with potentially widely varying outdoor levels not only from day to day or location to location but even by the time of day morning or afternoon. Further complicating classifications of homes as moldy or not based on ratio of indoor to outdoor samples, concentrations of mold/fungi in outdoor air vary in relation to many factors, including temperature, humidity, geography, ocean breeze, precipitation, season and others. The ratio of indoor to outdoor spore counts is related to the direct penetration of outdoor mold spores into indoor air via doors, windows, and other sites of air exchange. The indoor/outdoor ratio is strongly affected by the use and quality of a home s AC filter; and whether windows are open for much of the year as in California or almost never open as in South Florida. In several key papers studying the use and scientific validity of evaluating an indoor environment based on the ratio of indoor to outdoor spores counts, Spicer and Gangloff 3,4 have found: the levels of fungi in the outdoor air varied considerably between morning and afternoon with no pattern by species, time of day or location. numerically based criteria cannot consistently identify differences in indoor and outdoor data sets, and therefore are not adequately protective of public health, despite their intuitive appeal. Similarly, potentially misleading indicators of a problematic indoor environment through high false positive rates for numerical criteria potentially drive additional sampling and/or remediation that otherwise would not be indicated. Clearly there are limitations to using outdoor air comparisons to determine a home s moldiness level. In 2002 the EPA s Dr. Haugland and Dr. Vesper 5,6,7 in conjunction with U.S. Housing and Urban Development patented a procedure for analyzing mold using DNA profiling called ERMI Environmental Relative Moldiness Index. ERMI is not based on taking outdoor samples to be used to determine if a home is moldy or not. Instead ERMI relies on a national database of moldy versus not moldy homes used as a reference to determine moldiness or not and is based on studies of 1,096 homes across the U.S. as part of the 2006 HUD American Healthy Home Survey 8. Individual indices, ranked from lowest to highest were used to create the national Environmental Relative Moldiness Index (ERMI) Scale. A high index value means that there is or has been water damage that resulted in elevated levels of water indicator molds. A low value means less or no such problems. www.mold-toxins.com 2/28/2015 Page 4

It is widely argued that numerical values from traditional mold testing (culture and/or by analyzing mold spore characteristics under a microscope) cannot be reliably used to determine if a home has elevated mold or not. See Traditional mould analysis compared to a DNA-based method of mould analysis 9 by Stephen Vesper for a discussion of the limitations of traditional mold analysis. The question we ask here: Can ERMI do any better? Can ERMI testing be reliably used to determine if a home or office has problems or does not have problems? This is a very important question as mold assessors/remediators as well as physicians are prescribing ERMI testing to do just that determine if a home has current problems or not. And based on the outcome of a single ERMI test, decisions on remediation or health treatments are being made. How Does ERMI Work? For an excellent summary of ERMI, how it works and why it was developed see: Dr.Vesper s presentation at: http://www.hud.gov/offices/lead/nhhc/presentations/r-12_mold- Moisture_Assessment_Research.pdf ERMI was developed to measure a history of water damage by analyzing mold in carpet dust samples (principally in low income residences.) The relative moldiness of carpet dust is based on a comparison of 26 known water-damage indicator molds (Group 1) to 10 molds typically found outside and not water indicator molds (Group 2). The higher ERMI value (Group 1 versus Group 2) the more moldy/ water damaged a home. High ERMI values have been linked to a significant increase in the development of childhood asthma 5,6,7. Does ERMI Carpet Dust Testing Relate to Current Exposure? Currently most of the major commercial analytical labs that do mold testing offer ERMI analysis for both consumers as well as mold professionals. However the EPA has found that ERMI procedures are not meant to be a consumer product but meant solely as a tool for professionals. See the August 2013 EPA report from the Office of the Inspector General: http://www.epa.gov/oig/ reports/2013/20130822-13-p-0356.pdf. The Inspector General has concluded that: 1.) ERMI has not been peer reviewed or validated for public use. 2.) There is the potential for firms or individuals to overstate the implications of the ERMI tool results to clients in order to persuade them to undertake more costly remediation services. We agree with the Inspector General s conclusions for the following reason: ERMI was calibrated for use with carpet dust to correlate the history of water damage with the development of childhood asthma. And it works well for this purpose. ERMI was NOT developed to determine if current conditions are problematic or not. And ERMI does NOT do a good job evaluating current conditions of moldiness as we shall see. Mold in carpet dust may be 10 years old or older. Whether there was a water problem 10 years ago that was not properly remediated and resulted in children developing asthma, even though this is quite important to researchers, this is not what a typical mold investigator or home owner is interested in. www.mold-toxins.com 2/28/2015 Page 5

What a mold investigator or home owners/tenants/landlord wants to know is more limited and quite specific: Does a home currently have a significant mold problem or not? If so, what is the cause, location and the extent of the problem for the purpose of getting mold remediation quotes and fixing the moisture source that is causing the mold growth? What a physician or home owner/tenant/landlord and perhaps attorney wants to know as well: Is there significant exposure from the mold? What types of molds (mold toxins and/or allergens) is the occupant being exposed to? Current exposure to mold is due to breathing mold spores and mold particles in the indoor air. Mold in carpet dust does not necessarily correspond well to current exposure. Nor does analyzing mold in carpet or floor dust indicate the cause, location or extent of any current problems. But what it will do, measuring mold in carpet dust will often overestimate mold problems in a home. Why? When the carpet is old and/or the occupants are poor and can t afford to clean carpets after water exposure, carpets can build up a history of mold problems but will not significantly impact the current indoor air quality. There may not be any current problems or current exposure from mold but at the same time ERMI analysis of mold in carpet dust can be high or even off the chart. As a result, home owners can be persuaded to undertake more costly remediation services as noted by the EPA Inspector General and may also be persuaded to undertake unnecessary medical treatment for exposure to indoor mold. (Taking dust samples from refrigerator coils is even more problematic. Quite often there are minor mold and water damage problems in the sink cabinet which do not relate to mold exposure. The refrigerator fan may be pulling air from this otherwise contained area and massively overestimating exposure levels.) How Often Does ERMI Testing Result in False Positives? Of particular concern is how often ERMI gets it wrong and indicates mold problems when there are none. As a result (of this false positive) there would be unnecessary mold remediation and/or perhaps unnecessary medical treatment. We went back to the original Healthy Homes Study 8 : Correlation between ERMI values and other moisture and mold assessments of homes in the American Healthy Homes Survey by Vesper S 1, McKinstry C, Cox D, Dewalt G. In the Healthy Homes Study which is the basis for the acceptance of ERMI: Homes in the highest ERMI quartile were in agreement with visual inspection and/or occupant assessment [ONLY] 48% of the time but failed to detect the mold in 52% of the fourth quartile homes. In about 7% of lowest ERMI quartile homes, the inspection and occupant assessments overestimated the mold problem. Such percentages are not good at all. How can one expect to make a decision to remediate or not or to obtain medical treatment or not when half the time the high ERMI findings do not agree with the visual assessment. Of course the paper assumes that ERMI is right and the mold assessor is wrong. But how do we know that. We do not! Note that ERMI is ranked in Quartiles Quartile 1 is lowest ERMI score. Quartile 4 is the highest. www.mold-toxins.com 2/28/2015 Page 6

In the Healthy Homes Study which is the basis for the acceptance of ERMI, from Table 1 shown below: Water problems per occupant = Does not correlate with ERMI. Musty smell per occupant = Slightly correlates with ERMI. Musty smell per inspector = Does not correlate with ERMI. Visible mold per inspector = Does not correlate with ERMI. This is like the emperor that had no clothes. EPA says high ERMI means water damage and associated illness and low does not. But the correlation is very weak to non-existent. So if according to the EPA studies if there is slight to no correlation with current water problems, current odor or visible mold why are so many people basing health treatment or mold remediation decisions on ERMI? Makes no sense. ERMI Study Performed in France with Dr. Vesper of the EPA An independent study using ERMI was performed in France with Dr. Vesper of the EPA as one of the authors 10. They also found poor correlation between ERMI and visible mold. Figure below shows Environmental Relative Moldiness Index values for 40 dwellings assembled from lowest to highest (black diamonds) and the corresponding visual estimates of fungal contamination (gray columns). www.mold-toxins.com 2/28/2015 Page 7

In this same study NO correlation was found between ERMI and measured mold in the air. Figure below shows Environmental Relative Moldiness Index (ERMI) values for 40 dwellings assembled from lowest to highest (black diamonds). The logs of the total fungal cell concentrations as colony forming units per m 3 air sampled for the living room (gray columns) and bedrooms (white columns) are shown corresponding to the ERMI values for each dwelling. Where is the correlation between ERMI and mold in the air? There is none. If an independent research group that includes Dr. Vesper of the EPA (Vesper is co-developer and holds the patent for ERMI) cannot find a correlation between mold in the air and ERMI, or visible mold and ERMI why are so many people basing health treatment or mold remediation decisions on ERMI? Makes no sense. But this does not necessarily mean that mold DNA analysis itself is problematic. After all, ERMI was developed to provide a historical record of mold problems in homes and to correlate such history of mold problems with the development of asthma. And it works well (at least appears to work well) for that purpose. Unfortunately people are using ERMI for analyzing current problems, something for which it was not designed for and for which it does not do a good job. What about other DNA based analysis techniques or derivatives of ERMI not based on analyzing carpet dust, such as: Air samples or Dust samples from air filters, and then performing the ERMI/ DNA analysis of 36 molds on them? These are called ERMI-like analyses. We will look at these DNA based techniques next. www.mold-toxins.com 2/28/2015 Page 8

ERMI/DNA Air and Air Filter Samples Because the EPA research was based on historical studies linking mold and water damage to childhood asthma, sampling carpet dust was ideal. Carpet dust often does contain a history of mold and water damage. And it makes sense that dirty carpet that is full of mold toxins and allergens can have a negative impact on children crawling on them. However in terms of evaluating current conditions, ERMI/DNA based mold analysis of carpet dust has little value as Dr. Vesper s own studies have shown. But this is to be expected since carpet dust can be 10 years old or more. Why should one expect carpet dust to correlate with current conditions? One should not expect it to and it does not. These same ERMI/DNA methods though can be used to measure mold that is present in the indoor air either directly by taking air samples or by sampling air filter dust. These approaches are being applied in homes, offices and in hospitals 11,12,13. Taking air samples for DNA analysis can take many hours 14 especially when collecting samples in hospitals where the goal is to rule out mold spores in the air. DNA air sampling will generally require two trips from the mold assessor (turn on pump and later collect pump and sample) which is often impractical or too big of an expense for homeowners. However the next best thing to taking ERMI/DNA air samples is to check the indoor air for mold by taking a dust sample of a home s air filter. This takes only one trip. Taking an ERMI/DNA dust sample from a home s air filter should certainly be much more representative of the air that occupants are breathing than sampling carpet dust. Sampling an air filter can be done by either cutting out a piece of the air filter or by using a dry Swiffer (dust magnet) and swabbing the filter. The only lab accepting these forms of ERMI samples that we know of is Mycometrics (www.mycometrics.com). (We have no affiliation with Mycometrics.) Studies of Air Filter Dust Sampling A research group at the University of Texas at Austin 15 (http://www.ce.utexas.edu/prof/siegel/papers/ noris_2011_filter_bio_ae.pdf) has evaluated HVAC filters as a sampling mechanism for indoor microbial growth. The benefit of evaluating air filters is that the accumulated dust on the filters should represent a short term history of airborne mold since the last filter change a few months. So this would not represent a multi-year history of mold problems as when analyzing carpet dust. Nor would it be an instantaneous measurement of indoor air at one location as with an air sample. Mold levels vary from room to room as well as time of day and a dust sample from an air filter has the advantage of being not only a short term history of mold in the air but also represents mold found throughout the house. The group s research found that mold spores in the air filters closely resembled the mold spores found in the air over the month long period during which they took indoor air samples. This is as one would expect. On the other hand, the mold found in the air filters (and mold found in air samples) did not correlate well with mold found in surface dust. This also should be no surprise as we now know the limitations of testing for mold in carpet or floor dust which do not represent current indoor air conditions. Their research has concluded that a high efficiency air filter located in an AC, operating at a great fraction of time (either cooling or with FAN=ON) can be used as a surrogate for long term air samples. Hold this thought for the moment. Next we look at air sampling using DNA/ERMI and then contrast with air filter dust sampling. www.mold-toxins.com 2/28/2015 Page 9

Recent NIH Study Using DNA Techniques for Analyzing Air Samples A recent NIH funded study (by Meklin, Vesper et al 16 ) was performed by taking air samples (not carpet dust and not air filter dust) and analyzing samples using DNA based methods. What they found was of interest. They found that only 4 species of mold were correlated between indoor and outdoor samples. This would imply that the basis of ERMI which is to subtract the ten Group 2 molds (outside molds) from Group 1 (water damage indicator molds) makes no sense if applied to air samples. Therefore doing an air sample and submitting it for ERMI analysis is not acceptable. You cannot do an ERMI analysis on an air sample. Nevertheless this does not mean one cannot take an air sample and send it in to be analyzed for the levels of the 36 mold species covered by ERMI testing (the ERMI mold panel). This can be quite useful. 1.) Air sampling for DNA analysis captures all mold spores as well as mold micro-particles and is not a function of the efficiency of the air filter which tends to capture more of the larger spores and fewer small spores and fewer of the much smaller mold micro-particles/ mold fragments. 2.) The air sample results are provided in terms of mold spores (actually mold spore equivalents which includes DNA from both spores and from mold micro-fragments) per cubic meter of air and as such can be directly correlated with traditional air sampling either culture or microscopy based spore counts which are also based on spores per cubic meter of air (spores/m 3 ). Our firm (Certified Mold Free Corp.) recommends taking ERMI/DNA air filter samples as well as DNA air samples in additional to traditional air samples as part of a complete mold assessment when there are legal and/or medical issues involved. We have a few concerns with the technology but for evaluating current rather than historical conditions, DNA air sampling and DNA air filter dust sampling often called ERMI-like testing are far superior to taking carpet or floor dust samples. Concerns with ERMI/DNA Air & Air Filter Testing There are many significant concerns with regard to DNA Air and/or Air Filter testing. Many or most of these concerns also apply to carpet or floor dust testing. Nevertheless ERMI-like air sampling or ERMI-like sampling of air filter dust can be useful tools in the hands of a professional mold assessor especially when performed in addition to and not instead of traditional mold testing. 1. Settling rates vary depending on particle size. The AC is a long way from most of the home. Therefore one would expect that large size spores such as Stachybotrys (toxic mold) spores would be underrepresented in an air or air filter sample taken at the AC unless there was a great deal of activity in the indoor environment. For the same reasons one would expect large spores such as Stachybotrys to be over represented in carpet dust and not indicative of the amount in the indoor air. 2. Quite often there are mold and water damage problems in the AC closet. Assuming a quality air filter, many or most of the problem mold spores will be trapped in the air filter before contaminating the home or office s indoor air. This would result in the air filter sample overestimating the amount of problem mold in the indoor environment. www.mold-toxins.com 2/28/2015 Page 10

3. The quality of air filters varies considerably. No analysis exists as to how air filter quality (MERV rating) may affect ERMI-like filter dust values. Smaller particles are not captured as efficiently by lower quality air filers as by high quality (high MERV value) air filters. Therefore filter sampling may underestimate the amount of small mold spores (or mold fragments) versus larger spores. 4. The reader should understand that any type of ERMI/DNA result whether carpet or floor dust or from an air filter or from fridge coils in no way represents the level of exposure. A high ERMI value can be found in homes that are perfectly healthy. The ratio of Group 1 molds to Groups 2 molds could be high but the actual amount of overall mold involved may be negligible in terms of impacting the quality of the indoor environment. Michael Pinto in his article A QUICK PRIMER ON THE PERILS OF USING ERMI SAMPLES FOR POST REMEDIATION VERIFICATION FOR MOLD PROJECTS does a good job of explaining this problem. His write up can be found at: http://www.wondermakers.com/portals/0/docs/ermi%20for%20post-remediation.is.pdf Below is an example that explains why a consumer relying on ERMI (or HERTSMI-2) values can lead to a false conclusion that there are problems in their home when there are none. www.mold-toxins.com 2/28/2015 Page 11

On the previous page we have a lab report from an air sample taken in the author s home. The sample was taken by the 2nd floor AC and also one was taken outside. I ran the high speed 15 liters/minute sampling pump for 10 minutes and collected 150 liters of air for each sample. When the lab reports the results, they always convert to spores per cubic meter of air which is spores per 1000 liters of air. This allows comparison between different collection methods as some people use low speed air pumps (5 liters per minute) others use high speed pumps (15 lpm) and at times samples are taken for 3, 5, 10 or 30 minutes. Converting to spores/m3 is called normalizing the data. Therefore the Raw Count is multiplied by 1000/150 to get Spores per m3 shown in the column to the right of the raw count. The lab has Concluded that the indoor air in Not Elevated. That is an easy call to make because there is essentially nothing in the air. The indoor air is about as perfect as can be (even though we have carpet upstairs 5 years old and the AC is 14 years old.) Why is the indoor air spore count so low? Because everyone in our home is mold sensitive so: We keep the house clean but the kid s rooms have clutter and dust. We keep the AC and ducting clean. Windows are never open (typical for South Florida). We run the AC s 24/7 with FAN=ON so that the home s air is constantly being filtered; and We use MERV 11 rated air filters which are allergen rated air filters suitable for removing all sizes of mold spores. Now let s take a look at the ERMI analysis of the dust on the air filter for the AC servicing that area shown on the right. Three pages of the ERMI report are inserted next. The first column is the ERMI 36 molds with Group 1 molds (supposedly water indicator molds) on top and Group 2 molds below (supposedly outside molds that are not water indicator molds.) The next column is called Clean: Davie FL. This column contains the results from air filter dust filter in place for about 4-5 months from which a 6 x 6 piece was cut out of our air filter and sent to Mycometrics for ERMI analysis. The result (16.37) is off the chart and in error. There are two important points to be made here. A. Clearly what the EPA/Vesper has classified to be outside mold is not outside mold in South Florida. We live next to the Everglades. Outside mold here is not the same as in Cleveland or California or other places. For the same reason the trees, plants, grasses and weeds are completely different in the South, West, North and East, the mold that grows outside is completely different. ERMI got this analysis completely wrong because outdoor molds vary dramatically from location to location. ERMI is fundamentally flawed because it assumes outdoor spore distributions are similar throughout the country when this is completely false. B. And ERMI only you gives the ratio of what they call indoor water damage molds to outdoor molds. The result does not tell you how much is there. As we know dose makes the poison. In this case what we are breathing in our home is pure air that has essentially no mold contaminants. But because the air filter was capturing mold spores that trickle in the house whenever a door is open but this is going on for months we get a heavily moldy filter that has nothing to do with the actual day to day indoor air. ERMI got this analysis completely wrong because ERMI is not quantitative. www.mold-toxins.com 2/28/2015 Page 12

MYCOMETRICS, LLC. 11 Deer Park Drive, Suite 210 Monmouth Junction, NJ 08852-1923 Tel: 732-355-9018 ~ Fax: 732-658-5185 ~ Email: quest@mycometrics.com ~ Website: www.mycometrics.com Client: Dr. Gary Rosen Address: 2881 W Lake Vista Cir, Davie, FL 33328 Project ID: R&D Clean vs Moldy AC Filters Date Sampled: 2/6/2015 Date Received: February 9, 2015 Samples Submitted By: Gary Rosen Date Analysis Completed: February 13, 2015 Mycometrics Report No.: 150209-014 Mycometrics Report Date: 02/13/2015 ERMI Report This report includes the results of an air filter from our home (Clean: Davie FL) and an air filter taken from a mold infested home (in Tamarac FL.) www.mold-toxins.com 2/28/2015 Page 13

Air filter from our home. Note that the column on the right is the results from an air filter taken from a heavily contaminated, water damaged (Foreclosed) property. It is also off the chart but in this case there s plenty of mold. Location Clean: Davie FL Dirty: Tamarac FL Spore E./mg Spore E./mg Fungal ID \ Sample ID ERMI-Rosen-Clean AC filter ERMI-Rosen-Dirty AC filter Aspergillus flavus/oryzae 5 40 Aspergillus fumigatus 31 200 Aspergillus niger 180 12000 Aspergillus ochraceus 2 15 Aspergillus penicillioides 630 2100000 Aspergillus restrictus* 6 670 Aspergillus sclerotiorum 6 290 Aspergillus sydowii 140 20000 Aspergillus unguis 3 8700 Aspergillus versicolor 180 1500 Aureobasidium pullulans 100 69 Chaetomium globosum 72 320 Cladosporium sphaerospermum 210 2300 Eurotium (Asp.) amstelodami* 47 5200 Paecilomyces variotii 72 12000 Penicillium brevicompactum 1 25 Penicillium corylophilum 2 99 Penicillium crustosum* 1300 Penicillium purpurogenum 1 21 Penicillium spinulosum* Penicillium variabile 110 20000 Scopulariopsis brevicaulis/fusca <1 18 Scopulariopsis chartarum 19 13 Stachybotrys chartarum 8 23000 Trichoderma viride* 4 4 Wallemia sebi 58 27000 Sum of the Logs (Group I): 30.59 71.59 Acremonium strictum 1 5 Alternaria alternata 3 5 Aspergillus ustus 460 47000 Cladosporium cladosporioides 1 110 1300 Cladosporium cladosporioides 2 26 93 Cladosporium herbarum 57 98 Epicoccum nigrum 96 280 Mucor amphibiorum* 92 160 Penicillium chrysogenum 28 57000 Rhizopus stolonifer 3 930 Sum of the Logs (Group II): 14.22 25.52 ERMI (Group I - Group II): 16.37 46.07 www.mold-toxins.com 2/28/2015 Page 14

The sample(s) in this report was/were received in acceptable conditions. The results in this report apply only to the sample(s) tested. ** Concentration is rounded to two significant digits. Concentration is in Spore E./Sample if sample amount/area is NA. Spore E.: Spore equivalents; A spore equivalent may reflect the presence of any other fungal structures (i.e. mycelia) containing the same number of target genes as a spore. : Not detected within 40 cycles of PCR amplification. * Genetically closed-related species may be detected in the indicator assay: Eurotium (Asp.) amstelodami covers E. chevalieri, E. herbariorum, E. rubrum and E. repens. Penicillium spinulosum covers P. glabrum, P. lividum, P. pupurescens, and P. thomii. Trichoderma viride covers T. koningii and T. atroviride. Aspergillus restrictus covers A. caesillus and A. conicus. Mucor amphibiorum covers M. circinelloides, M. hiemalis, M. indicus, M. mucedo, M. racemosus, M. ramosissimus and Rhizopus zygosporus, R. homothalicus, R. microsporus, R. oligosporus, R. oryzae. Penicillium crustosum covers P. camembertii, P. commune, P. echinulatum, P. solitum. Approved by: King-Teh Lin, Ph.D. Microbiologist www.mold-toxins.com 2/28/2015 Page 15

5. Independent research 17 has found: Efficiencies associated with recovery of DNA from aerosol filters were low, and excluding these efficiencies in quantitative analysis led to underestimating the true aerosol concentration by 10 to 24 times. 6. False negatives with DNA testing 18 : Independent research has found in some samples, Asp. versicolor was detected by culturing, but not by qpcr [DNA]. The reason for inefficient detection of Asp. versicolor is unclear. (An issue with all forms of ERMI sampling.) 7. The ERMI index is based on samples taken in cities. The outside air in cities can be completely different from the outside air in the suburbs or country. So what floats into a home from the outside which is the basis of what ERMI considers normal depends on the location (city, suburbs, or country.) Depends on whether you are close to wet lands (big issue in S Florida) or not. Our firm has found strongly false positive results for ERMI tests taken in the suburbs near wetlands. 8. Studies have shown that more mold fragments are released from mold than mold spores. Because fungal fragments have DNA, ERMI/ DNA analysis will significantly overestimate mold spore contamination because it does not distinguish mold spores from mold fragments (mold micro-particles). See chart to the right. On the other hand, since fungal particles and not only spores can cause health problems, ERMI/DNA s being able to detect both spores and fragments can be a major advantage over traditional methods of mold assessment for which the sub-micron (tiny) fungal fragments are invisible and not detected. 9. Furthermore as we have seen in concern #1, the settling rate of mold fragments is in days and not in minutes as is the case for large spores such as Stachybotrys or Chaetomium spores. The quality of the air filter will again come into play. How much of the captured mold DNA material is from mold fragments that will stay suspended in the air for long periods of time and how much from intact mold spores? Depends on the filter quality. Depends on how far the AC return air is from the source of the mold contamination. Mold DNA in air filters may significantly over-estimate spore counts (heavier particles) versus the lighter mold fragments. 10. Respiratory deposition is a function of the size of the mold contaminants. Most larger whole spores are deposited in the upper respiratory airways and never reach the lungs (see chart on right). On the other hand, mold fragments tend to be respirable and a significant number will reach the alveoli where they must be engulfed by macrophages for removal. Removal by macrophages is a slow process and respirable mold particles result in higher levels of mold toxin exposure than large mold spores that are cleared from the upper airways and deposited in the gut for removal in the stool. Settled floor or carpet dust for ERMI/DNA analysis may underestimate respiratory deposition because most of the small airborne mold micro-particles remain airborne for long periods of time and will be removed by a home s air filter rather than settle in floor dust. 11. ERMI/DNA methods do not distinguish viable from non-viable spores. From the extensive review article Fungal spores: A critical review of the toxicological and epidemiological evidence as a basis for occupational exposure limit setting 19 by Wijnand Eduard: The increased production of allergens in the germination phase suggests that viable spores may be more allergenic than dead spores if germination occurs in the respiratory tract. www.mold-toxins.com 2/28/2015 Page 16

Studies have shown 20 that approximately 99% of the mold spores detected by ERMI are nonviable. For this reason, methods such as ERMI or HERTSMI-2 that do not distinguish between viable and non-viable spores will always significantly overestimate the potential for allergic response. 12. Outdoor levels vary by four orders of magnitude between regions and seasons over a range of <20 to >10 5 cfu/m 3 ; see review by Gots et al 21 (2003). Outdoor levels are highest in warm regions and in the summer and autumn in temperate regions, and lowest in the winter or on the coast with onshore breeze. In light of the extreme variation in outdoor spore levels, can one always agree with the main assumption of ERMI that outdoor count variation can be corrected for by subtracting out Group 2 values? This has never been evaluated over a wide range of outdoor levels. Refer again to the Air Filter ERMI analysis from my house showing that outside molds vary dramatically from geography to geography making ERMI inherently inaccurate. Steven Parkhurst has concerns with levels of Group 2 (outside molds) in his article CAN ERMI BE IMPROVED? (http://www.experts.com/articles/mold-dna-detection-can-ermi-be-improved-by- Steven-Parkhurst) In Vesper's initial study, normal Group 2 log sums ranged from 7 to 14. In my experience, however, Group 2 log sums often range from 2 to 5. The only way practitioners will know whether their Group 2 sums are low is for the lab to identify the problem on its ERMI report. Low Group 2 log sums result in artificially high ERMI values, making the home appear moldier than it is. This has significant ramifications on recommendations made to building owners and occupants. EPA tends to dismiss this phenomenon as infrequent and of no consequence; sometimes the agency suggests evaluation of Group 1 molds alone but does not mention how this impacts the ERMI score. In my year-plus experience with ERMI sampling, I can't recall a single sample where the ERMI score was "in the green"-that is, below minus-four, the 25th percentile or less on the moldiness index. My samples included those from new homes, homes where no water impact was found, and clearance on mold remediation projects. 13. Mold spores in most or all cases can induce some degree of inflammation in experimental studies 22. However, the response to mycotoxin-producing and pathogenic mold species is much stronger than to non-toxin producing molds that produce only inflammation. ERMI does not distinguish toxin producing molds such as Stachybotrys and Chaetomium from non-toxin producing molds and therefore will often underestimate the potential for health related problems from toxic mold exposure. 14. Asthmatic patients that are allergic to Penicillium sp. or Alternaria alternata, (or anyone allergic to Penicillium mold) are affected by vastly lower exposure levels to these allergenic molds than are non-allergic individuals. ERMI may therefore underestimate the potential for health related problems from mold exposure in such cases. 15. Due to the expense of ERMI analysis typically only one sample is taken. In the event of hidden mold this does not allow one to pinpoint the location of a problem. For instance, traditional air samples analyzed by microscopic analysis are low cost can be taken throughout a home. If elevated in one particular location (hot spot) that may indicate hidden mold. If on the other hand mold levels are elevated and similar throughout the home or throughout the zone of an AC unit in homes that have more than one AC, this will usually indicate AC, ducting or AC closet problems. Generally such testing is more useful in locating mold for the purpose of removal than using DNA based analysis methods. www.mold-toxins.com 2/28/2015 Page 17

Conclusions We concur with the EPA Inspector General: ERMI testing is not meant to be a consumer product. ERMI testing (based on analysis of carpet/floor dust or AC filter dust) as a stand alone tool is NOT suitable for determining if a home or office has current or even past mold problems. Decisions on remediation or health treatments should not be made based on the outcomes of a single or even multiple ERMI tests. Despite the numerous concerns and/or limitations, ERMI/DNA testing using air sampling or air filter sampling which unlike carpet dust evaluates current (not historical) indoor mold levels can be powerful, useful and generally reliable techniques in the hands of a knowledgeable mold professional. DNA Analysis of Air Sampling: While air sampling using DNA techniques cannot apply ERMI analysis to the sampling results as the ERMI index was developed for carpet and floor dusts, DNA based air sampling is the ONLY method to enable one to check for (rule out) both mold spores as well as mold micro-particles (mold fragments) at the time of inspection. DNA based air sampling has the benefit that it will collect 100% of all the mold in the air including both spores as well as mold micro-particles. Traditional mold sampling (either microscopy based spore counts or culture based) does not detect mold micro-particles. Mold micro-particles are completely invisible to mold spore counts or culture methods. But DNA air sampling because the sampling cartridge has a very small filter pore size (0.45 micron) will efficiently collect all spores as well as mold micro-particles and has a very useful place as a tool for the professional mold assessor especially when medical and/or legal issues are involved. ERMI lab results for carpet dust and air filter dust samples are given in spore equivalents per mg of dust. This can be called quantitative. But what does this compare to? How much is a lot and how much is a little in terms of spores per mg dust? No one knows what spores per mg dust means. On the other hand, DNA based air sampling results provide the indoor air concentration of mold (given in spores per cubic meter of air). Spores per cubic meter of air is the same reference standard that is used by traditional air sampling for providing air sample results. Every mold assessor knows what a small amount of mold or a large amount of mold is when it comes to traditional air samples where concentration is measured in terms of spores/m 3 as they have taken hundreds or thousands of samples with results given in spores/m 3. Therefore results of DNA air sampling has meaning in terms of how much is there. Is there elevated levels of mold in the air or not. Traditional microscopy based analysis of spores does not provide details about mold species for example cannot enumerate or distinguish between Penicillium or Aspergillus molds) and therefore provides limited information about what potential mold toxins are present. DNA air sample analysis provides this sort of detailed information and is also quantitative as it provides results in the familiar format spores/m 3 unlike dust samples where results are presented as spores/mg dust which has no real meaning to mold assessors or remediators. See air sample results on next page based by DNA analysis at Mycometrics. The analysis uses the ERMI panel of 36 molds but does not include breakdown into Group 1 and 2 molds or provide an ERMI value as ERMI analysis is for carpet and floor dusts and not air samples. We ve taken a large volume of air 7200 liters (8 hours sampling at 15 liters per minute). See column labelled Air Volume. www.mold-toxins.com 2/28/2015 Page 18

Air Sample MYCOMETRICS, LLC. Mycometrics Sample ID Air Volume (L) Fungal ID Spore Conc. ** Percentage Client Sample ID Equivalents (Spore E./m3) Location 150130-035-001 7200.0 Acremonium strictum ERMI-Air sample MBR closet Alternaria alternata XXXX Weston FL Aspergillus flavus/oryzae Aspergillus fumigatus Aspergillus niger 1 Aspergillus ochraceus 1 Aspergillus penicillioides 590,000 82,000 95.6% Aspergillus restrictus* 61 9 Aspergillus sclerotiorum 13 2 Aspergillus sydowii 100 14 Aspergillus unguis Aspergillus ustus 1 Aspergillus versicolor 210 29 Aureobasidium pullulans Chaetomium globosum Cladosporium cladosporioides 1 120 16 Cladosporium cladosporioides 2 10 1 Cladosporium herbarum Cladosporium sphaerospermum 190 26 Epicoccum nigrum 1 Eurotium (Asp.) amstelodami* 13,000 1,700 2.0% Mucor amphibiorum* Paecilomyces variotii Penicillium brevicompactum 5 1 Penicillium chrysogenum Penicillium corylophilum Penicillium crustosum* Penicillium purpurogenum Penicillium spinulosum* Penicillium variabile 4 Rhizopus stolonifer 2 Scopulariopsis brevicaulis/fusca Scopulariopsis chartarum 1 Stachybotrys chartarum Trichoderma viride* Wallemia sebi 14,000 2,000 2.3% Total: 85,798 100% The sample(s) in this report was/were received in acceptable conditions. The results in this report apply only to the sample(s) tested. ** Concentration is rounded to two significant digits. Concentration is in Spore E./Sample if sample amount/area is NA. Spore E.: Spore equivalents; A spore equivalent may reflect the presence of any other fungal structures (i.e. mycelia) containing the same number of target genes as a spore. : Not detected within 40 cycles of PCR amplification. * Genetically closed-related species may be detected in the indicator assay: Eurotium (Asp.) amstelodami covers E. chevalieri, E. herbariorum, E. rubrum and E. repens. Penicillium spinulosum covers P. glabrum, P. lividum, P. pupurescens, and P. thomii. www.mold-toxins.com 2/28/2015 Page 19 MycometricsReport No.: 150130-035 Page 2 of 3

Dust sample from Clothes Location XXXXX Weston FL Spore E./mg ERMI-Closet Dust Clothes Fungal ID \ Sample ID Aspergillus flavus/oryzae Aspergillus fumigatus Aspergillus niger 5 Aspergillus ochraceus 1 Aspergillus penicillioides 230000000 Aspergillus restrictus* 370 Aspergillus sclerotiorum 2 Aspergillus sydowii 11 Aspergillus unguis 2 Aspergillus versicolor 60 Aureobasidium pullulans 9 Chaetomium globosum Cladosporium sphaerospermum 76 Eurotium (Asp.) amstelodami* 110000 Paecilomyces variotii Penicillium brevicompactum <1 Penicillium corylophilum Penicillium crustosum* Penicillium purpurogenum Penicillium spinulosum* Penicillium variabile Scopulariopsis brevicaulis/fusca Scopulariopsis chartarum Stachybotrys chartarum 1 Trichoderma viride* Wallemia sebi 580000 Sum of the Logs (Group I): 28.68 Acremonium strictum Alternaria alternata Aspergillus ustus 1 Cladosporium cladosporioides 1 40 Cladosporium cladosporioides 2 1 Cladosporium herbarum 3 Epicoccum nigrum 51 Mucor amphibiorum* 1 Penicillium chrysogenum 1 Rhizopus stolonifer Sum of the Logs (Group II): 3.79 ERMI (Group I - Group II): 24.89 www.mold-toxins.com 2/28/2015 Page 20

Note that the concentration of mold in the air is provided: 85,798 spores per cubic meter of air (5th column). That s a lot of mold in the master bedroom closet. That s because mold was growing on clothes in the carpet due to elevated dampness in the house. See picture on the right. Following the air sample results we include the Mycometrics DNA analysis of the white dust on the wool jackets in the closet. Because this is a dust sample, the results are given in terms of ERMI with Group 1 and Group 2 molds along with ERMI value. Note that the molds in the air are the same as the molds on the clothes. Makes sense. What else do we get from the air sample report besides there s a whole heck of a lot of mold in the air. No wonder the kids are sick and have been hospitalized with alleged mold exposure. The DNA analysis provides both the genus and species of molds present. The highest concentration of mold present is Aspergillus penicillioides. Checking Wikipedia we find: Aspergillus penicillioides facilitates the growth of house dust mites such as Dermatophagoides pteronyssinus. Aspergillus penicillioides is a common indoor fungus in damp buildings where it has been associated with allergic rhini s. This fits with the kids having some kind of allergic reaction to mold and/or dust mites in this home. Also finding Aspergillus penicillioides and Eurotium present (and very little else) because these two molds are dry molds (xerophilic) tells us that at least in this part of the house these molds are most likely the result of elevated moisture in the air and not from water damage which would also produce water loving molds such as Stachybotrys and Chaetomium. We can therefore assume that the AC is not working properly and not efficiently dehumidifying the air; or the occupants are leaving the windows open when running the AC; or they have the AC set at too high a temperature to save $$ on cooling costs. In this case the AC and ducting were badly mold contaminated and not operating properly. See picture of heavy mold and dirt on the interior lining of the air handler chassis on the right. This is all powerful information in the hands of a professional mold assessor/remediator and useful information for the client s physician as well. www.mold-toxins.com 2/28/2015 Page 21

DNA Analysis of Air Filter Dust: While the ERMI index developed for carpet dust will not make complete sense when used for analyzing air filter dust there can be usefulness in performing an ERMI analysis on air filter dust as this procedure will provide an excellent short term history with detailed species breakdowns of what has been in the indoor air and breathed by occupants. See pages 13-15 of this report. In the hands of a professional it may be useful to have the results broken down into Group 1 and Group 2 molds. But don t put too much faith in the actual ERMI value. Beware the ERMI or HERTSMI-2 analysis of air filter dust is not a stand alone test and must be performed by a mold professional along with visual inspection and collecting samples for traditional analysis. DNA analysis of air filters, unless the air filter is super high quality and captures ultra fine particles, will underestimate the amount of mold micro-particles in the air whereas DNA based air samples represent 100% of both spores and particles in the air (in the sample area at sample time.) However, the main issue with DNA analysis of air filters versus DNA based air samples is that the air filter dust results are given in spores/mg dust. What does that mean in terms of the concentration of mold spores and particles in the indoor? What does this mean in terms of exposure? Nothing actually. As and example, I have a client that is mold sensitive and looking for a home to rent that is free of elevated mold. He sent me the lab result below of an air filter that he sampled using a Swiffer (using the limited HERTSMI-2 panel rather than the 36 mold ERMI panel.) He wanted to know based on the result if the home was safe to rent. I told him that I had no idea because the results do not show how much mold is in the air just how much is on the air filter. Of course his doctor who he trusts with his health was the one that recommended he take his own samples and send them to Mycometrics for analysis, so I assume that he thought I was an idiot for not being able to interpret the results. After all, he took a sample of the home s air filter as recommended by his physician to determine if the home was safe or not. How could I be so dense. In this case I recommended that he put the current owners of the home up for the night in a nice hotel and spend the night there and see if he is irritated or not before signing the lease. I did not hear back from him. There is useful perhaps information in this HERTSMI-2 sample. There s Chaetomium present which probably means that there was some recent mold problems in the home as this is a water damage indicator but the home is on a swampy kind of lake and maybe this just blew in from the outside. A consumer taking such tests not a good idea. We agree with the EPA Inspector General. Such testing should be in the hands of a mold professional and not a consumer or health professional. www.mold-toxins.com 2/28/2015 Page 22

Multiple Methods for Mold Assessment Recommended: A combination of air filter DNA testing; air sampling with DNA analysis (4 hour sample time at 15 liters/minute air flow rate minimum); along with traditional air sampling throughout the home is recommended if one is concerned about identifying hidden mold problems or thoroughly characterizing the indoor mold for legal or medical requirements. Traditional air sampling should include collecting spores for both: Microscopic analysis which includes both dead and alive (viable) spores + Culture analysis to detect (or rather estimate) the percent of the detected mold that is viable which is newly produced from living mold and therefore more allergenic than dead (non-viable) mold. Testing should always be accompanied by visual inspection of mold and water damage along with inspection of the AC, ducting and AC closet as well as mold odor assessment. Occupants should be asked which areas are most irritating as well as describe any mold related symptoms they may have. For mold health effects, see Guidance for Clinicians on the Recognition and Management of Health Effects related to Mold Exposure and Moisture Indoors Appendix A Table A (Sept 2004) on the U.S. EPA Web site: http://doem.uchc.edu/consultation_outreach/indoor_environments/pdfs/mold_guide.pdf For further readings on some of the concerns with ERMI, along with benefits of ERMI see 23,24,25,26,27,28 Note about mold testing: We agree with the EPA when they question what may be the overuse of mold testing. When there is visible mold, testing is not typically useful and resources may best be spent removing mold rather than characterizing mold. See EPA mold guidelines (http://www.epa.gov/mold/moldguide.html) www.mold-toxins.com 2/28/2015 Page 23