Eighty-One Errors Made by the IOM Committee in Its Report

On March 15, 2013, the Institute of Medicine (IOM) published "Review of the Department of Labor's Site Exposure Matrix Database."

The second of four chapters in the report is "Haz-Map Database."

The third chapter "Site Exposure Matrix Database" contains two tables that are very misleading.

Those two tables are corrected and shown with a blue heading in the following two sections.

A total of 26 errors were found in these two tables (red heading)

The last two sections on this page are "Typographical Errors" and "Conceptual Errors."

Errors made by the IOM committee = 26 (Tables 3-2, 3-4) + 12 (typos) + 43 (conceptual) = 81 total errors

Errors in Haz-Map found by the committee = 4 (These 4 errors have been corrected in Haz-Map.)



15 Errors in Table 3-2

Table 3-2 Selected Missing Links for Toxic Substance-Cancer Based on Sufficient Evidence of Cancer in Humans (Revised)

(These agent-disease links have been added to Haz-Map and will be published later this year.)

SEM Substance Cancer Site
Asbestos Ovary (established by IARC in 2012)
Hepatitis B Virus (Chronic Infection) Liver
*Hepatitis C Virus (Chronic Infection) Liver
Plutonium Liver

*The IOM Report failed to list Hepatitis C virus as an occupational cause of liver cancer.


Table 3-2 Errors in IOM Report

Not Occupational Cancers

SEM Substance Cancer Site
Arsenic Urinary Bladder (Drinking water; not occupational); See Bladder Cancer.
Iodine-131 Thyroid (Children; not occupational); See Thyroid Cancer.
Thorium Multiple Sites (IV injection of medical contrast agent; not occupational); See IARC Monograph 100D.


Table 3-2 

Why did the committee call these "missing" when they knew Haz-Map and SEM were being updated in December 2012? 

These are all 2012 IARC changes. To see update schedule, go to the table at the bottom of this page.

SEM Substance Cancer Site
1,3-Butadiene Leukemia; Lymphoma, Non-Hodgkin
Coal tar pitch volatiles Lung
Diesel exhaust Lung
Formaldehyde Leukemia
o-Toluidine Urinary Bladder


Table 3-2 Radioactive Elements

(These links are in Haz-Map. They are not in SEM because Part E does not cover radiation-induced cancer.)

SEM Substance Cancer Site
Plutonium Bone; *Lung
Radon Lung
Radium Bone; Paranasal sinuses
Strontium-90 See occupational cancer links for "Radiation, ionizing."

*The IOM Report failed to list the established link between plutonium and occupational lung cancer.


Table 3-2 Not established human carcinogen

SEM Substance Cancer Site
Trichloroethylene Kidney



11 Errors in Table 3-4

Table 3-4 Selected Missing Toxic Substance-Noncancer Disease Links Based on Evaluations by Authoritative Organizations (Revised)

(This agent-disease link has been added to Haz-Map and will be published later this year.)


SEM Substance Human Disease Authoritative Organization
Welding fumes *Metal fume fever DOL, 1989; IARC 1990; NIOSH, 1988

Note that the disease is in Haz-Map and it is linked to zinc and other metals, but it was not linked to the agent "Welding fumes (not otherwise specified)." Welding on most metals will not cause metal fume fever, but welding on galvanized (zinc-coated) metal can cause metal fume fever.

Table 3-4 Errors in IOM Report

Potential Adverse Effects of Chemicals But Not Occupational Diseases


SEM Substance Adverse Effects Reported in Humans Established Occupational Diseases
Antimony Cardiovascular Pneumoconiosis
Carbon disulfide Cardiovascular Carbon disulfide, chronic toxic effect
Carbon monoxide Cardiovascular Asphyxiation, chemical
Chromium VI Male Reproduction Not occupational agent that reduces male fecundity; [Rosenstock, Table 27.2.8]
Dibutyl Phthalate Male Reproduction Not occupational agent that reduces male fecundity; [Rosenstock, Table 27.2.8]
2,4 and 2,6 Dinitrotoluene Cardiovascular Allergic contact dermatitis;
Hydrogen Cyanide CNS and Endocrine Asphyxiation, chemical
Methylene chloride Cardiovascular Solvent, acute toxic effect; Encephalopathy, chronic solvent
Rotenone A few cases of peripheral neuropathy None; May cause irritation;
Tetrachloroethylene Central Nervous System (CNS) Solvent, acute toxic effect; Encephalopathy, chronic solvent
Toluene Central Nervous System (CNS) Solvent, acute toxic effect; Encephalopathy, chronic solvent

Just because a chemical has the potential to cause an adverse effect does not mean that there exists an established, preventable occupational disease (with an ICD-9 code) associated with that chemical. For further information about adverse effects of particular chemicals, see the specific chemical profiles and the section on More Research Needed.



12 Typographical Errors

Page/Paragraph Typographical Error
5/3 14 DOE facilities {114 DOE facilities}
37/Box 2-2 "Experiment" 2 sentences separated by a comma;
37/Box 2-2 "Analogy" 2 sentences separated by a comma;
37/1 Dangling phrase ", of bias."
43/1 The word "however" in the last sentence of the first paragraph does not make sense in this sentence.
51/1 contactors {contractors}
54/4 Haz-Map,which {Haz-Map, which}
67/4 Should be period after "database" and then "For example";
76/2 It should be "quantitative information about the potency of a substance."
Table 3-3 "Liver (cholangiocarcinoma) for hepatitis B virus" already in previous table;
103/3 president {President}
119/Toluene 08-88-3 {108-88-3}



43 Conceptual Errors





*Informed  Jan/2013
4/1 Job tasks (the industrial hygiene perspective); Job tasks (the epidemiology perspective); Yes
4/1 Periodically updated; Continuously (almost daily) updated; Published quarterly; Yes
4/2 Lack of transparency in data sources; Data sources listed and methods published in AJIM; Yes
4/2 Lack of weight-of-evidence criteria for agent-disease links; See response to question #1 in 2nd Meeting Questions. Yes
4/2 Overreliance on textbooks; This is an opinion. What is the evidence?  
4/2 Information may be neither comprehensive nor up to date; No examples given;  
5/1 Has ambiguous criteria for identifying agent-disease links; Ignored my presentations of SHE(O)s & DiseasesbyCat.doc; See slides 15-18 in 1st Meeting Presentation. Yes
6/1 Delays in updating links in Haz-Map and thus in SEM; See p. 63, paragraph 2: ". . . this is an acceptable time frame for updates."  
6/3 Weight-of-evidence approach critical; Already used in Haz-Map for cancers; The best attempt to use this approach for noncancer diseases was the SHE(O)s published by NIOSH in 1991, which is the foundation and precursor of Haz-Map Yes
6/3 Inclusion of citations would enhance utility; The DOL response to the IOM report says, "The Haz-Map database does contain source references to each link that is in that database;" Yes
6/3 Peer review increases public confidence; Probably the most used database of chemicals is the NIOSH Pocket Guide. It is not peer reviewed. Neither is the ACGIH Documentation of TLVs and BEIs.  
6/3 Peer review increases accuracy & comprehensiveness; There are many examples of peer-reviewed papers of poor quality.  
6/3 Peer review ensures current information; This is not necessarily true, e.g., if the reviewers meet only once a year.  
7/1 TOXLINE, IRIS, and NTP would supplement Haz-Map. These databases already used, especially HSDB; Haz-Map includes all relevant chemicals in HSDB. IRIS and NTP cover only a few hundred chemicals that are also covered in better databases like ATSDR and ACGIH.  
7/3 Use of references in Haz-Map not systematic/comprehensive; Any examples? What is the evidence for this opinion?  
7/3 NTP toxicology reports not used at all; Many references to NTP; Haz-Map includes NTP cancer designations (not published by NLM but in ExPub version); Yes
7/3 OEHHA is a good resource for occupational toxicology; According to the department description accessed from the home page, its mission is to assess health risks of environmental contaminants. It works with the Dept. of Pesticide Regulation to protect applicators. No other occupational toxicology mentioned;  
7/3-8/1 These experts use weight-of-evidence approach to conclude toxicity; This is done for cancer only. For noncancer diseases, I see no such approach in the most important databases: NIOSH, ACGIH, ATSDR, CAMEO, CHRIS, HSDB, ICSC, IDLH, and EXTOXNET.  
8/3 Not transparent because citations to evidence not included; The DOL response to the IOM report says, "The Haz-Map database does contain source references to each link that is in that database;" Yes
8/4 Quality-control review will ensure accuracy; Inspection does not ensure quality; See page by Institute of Healthcare Improvement.  
12/2 It is difficult to determine the evidence base for the noncancer agent/disease links. See the second and third paragraphs of my "Summary" on page 4 of Jan. 2013 response. Yes
12/3 There is no peer-review of Haz-Map. See the first paragraph of my "Summary." See slide #34 in 1st Meeting Presentation Yes
25/2 Author is board certified in occupational medicine; No mention of MPH degree or 2 year fellowship in occupational medicine; Yes
Table 2-1 Adverse Effects Names of adverse effects are mixed up with categories of adverse effects. Yes
31/2 REPROTOX not available online; Available online by subscription; Of the 36 online books and databases, 29 are free. Yes
31/4 No transparent selection criteria for journals used in review; There is a list of journals used for each review. The ones selected contained the most useful articles in the 2002 journal review. Yes
42/3 No details on what constitutes a consensus; See details on my website pages: "Bronchitis, chronic" and "Renal Diseases." Yes
42/3 No details on what evidence reviewed in making links; Did you read the Haz-Map website? See Pneumoconioses, Toxic Pneumonitis, Pulmonary FibrosisOccupational Asthma, Allergic Contact Dermatitis, Toxic Hepatitis, and Methemoglobinemia. Yes
42/3 No details on good industrial hygiene; Each Haz-Map agent is linked to any of 54 industrial processes in which exposure may occur. Each disease is linked to any of 227 high-risk job tasks in which sufficient exposure may occur to cause the disease. Yes
43/1 No rules to determine if sufficient evidence for causal links; This was explained in the 2nd and 3rd paragraphs of my "Summary" in the Jan. 2013 response. What rules are needed to link lead to lead poisoning or carbon monoxide to chemical asphyxiation? See slide # 23 Metals and Occupational Diseases. Yes
44/3 Does not consider animal & mechanistic data; Of course, all toxicology data is considered, but a chemical that causes disease in animal ingestion experiments does not necessarily cause disease in workers exposed by skin or inhalation routes. Yes
46/2 Haz-Map not reviewed by anyone at NLM for accuracy, bias, and comprehensiveness; If true, then why did the department that publishes TOXNET invite me to publish Haz-Map? See slide #35 in 1st Meeting Presentation Yes
46/5 Haz-Map relies heavily on 3 textbooks. Why does the committee mention only 3 books when the content is loaded with references from many journals, databases, and over 20 main textbooks? Yes
47/1 Haz-Map cites only IARC for cancer designations. Haz-Map also includes ACGIH and NTP. These 2 fields are not yet published by NLM but they are published in the ExPub version of Haz-Map. Yes
47/2 References are not accessible. Of the 36 online books and databases, 29 are free. Of my 21 main textbooks, all are for sale on amazon.com (April 3, 2013). Yes
77/2 IOM-recommended sources use weight-of-evidence approach; Yes, for cancer; No such approach for noncancer diseases is used in ACGIH, ATSDR, NIOSH, or OSHA.   
Table 3-1 The NIOSH Pocket Guide is an authoritative source. This book has no references! Henry Chan, CIH was the technical editor. It was first printed in 1978. "The data were collected from a variety of sources . . ." According to the website, it is updated periodically.  
81/3 IARC Group 1 is only criteria used by Haz-Map for determining cancer links. The committee acknowledges on page 43, paragraph 2 that the "Occupation" chapter by Siemiatycki et al. is also used. A link requires Group 1 + occupational exposure. Yes
Table 3-3 Shows "Chloramphenicol," an antibiotic, as causing disease; Could cause leukemia by ingestion or injection, but not after occupational exposure;  
87/4 Occupational cardiovascular disease not clear in Haz-Map; See More Research Needed.  
88/1 These CNS effects are different from "encephalopathy, chronic solvent." According to "Solvent Neurotoxicity" by FD Dick, "Solvents, such as styrene, may cause sub-clinical colour vision losses." See PMID: 16497867 for free full-text article.  
96/3 Haz-Map lacks clear weight-of-the evidence approach. What weight-of-evidence approach did the committee use to make Table 3-4?  
100/1 NTP OHAT toxicology reports are an important resource absent from Haz-Map. The website on April 3, 2013 showed only 20 chemical profiles of which 3 were drugs.  

*This is my Jan. 2013 response to the 15-page "Haz-Map Info for Brown.doc." I received the 15 pages from IOM on Jan. 4, 2013. I was given 4 days including Saturday and Sunday to respond to the document written after 18 months of work by 12 primary reviewers, 5 members of IOM staff, 7 secondary reviewers, and 2 oversight reviewers.


Revised: July 16, 2013


All rights reserved. Haz-Map® 2017