Indoor Air Quality

Report Title: Investigation of Health Concerns at Madison Area Technical College, Truax Campus

Drafted by John S Hausbeck, Environmental Epidemiologist, Madison Department of Public Health

(Note: this report is posted online by the MATC Environmental Health and Safety Office)

Abstract

A long history of indoor air quality complaints and health concerns exists at the Madison Area Technical College (MATC) Truax campus. While ongoing efforts to correct identified ventilation and other building-related issues are progressing, there continues to be concern that excess levels of illness or poor health exists among occupants of the main building on the MATC Truax campus. The purpose of this study was to evaluate the health status of full-time faculty and staff at the MATC Truax campus to determine if levels of illness or symptoms throughout campus exceed expected levels. For most illnesses included in this study, levels observed at MATC Truax did not exceed levels observed in referent populations. However, levels of migraine (c²=13.84, p=0.001) and sinus infection (c²=48.47, p=0.000) were found to be significantly greater than local referent populations. No symptoms or groups of related symptoms were found to be significantly greater than levels reported for non-problem buildings. Pearson correlation analysis did not find levels of illness or symptoms, including migraine and sinus infection, to be strongly correlated with any of the personal, workplace, or environmental conditions reported in this study. This significant increase in migraine or sinus infection does not imply a cause and effect relationship with the MATC Truax campus. Additional study is required before this relationship could be determined. This study also supports the recommendations that MATC Truax administration continue to address the environmental issues identified. Also, all MATC Truax occupants, administrators, faculty, staff, and students alike, need to ensure that indoor air quality issues are communicated openly and accurately.

Introduction

MATC Truax campus was first occupied in October of 1986 shortly after the majority of construction was completed. Since that time there have been many complaints concerning the indoor air quality in the main building at on the MATC Truax campus. In 1998, Ald Santiago Rosas requested that the Madison Department of Public Health assess the status of the indoor air quality issue at MATC. After a review of existing documentation and discussion with MATC faculty/staff and other state and local agencies, it was learned that the MATC Truax facility met the requirements of the WI building code; however, building occupants continued to report concerns about chemical odors, stuffiness, building cleanliness, and physical discomfort. Some individuals also reported that these conditions were causing various health effects including headache; eye, nose and throat irritation; fatigue; and respiratory illness. In response to these concerns, MATC administrators hired a consultant in the early 1990's to evaluate the building heating and ventilating system and make recommendations to correct identified problems with the system. Since receipt of the consultant's report, MATC staff have been working to implement these recommendations. However, concerns that faculty, staff and students are becoming ill as a result of exposure to the MATC Truax environment have continued. Ongoing health concerns include many of the same illnesses and symptoms as reported in the past with the addition of sarcoidosis.

The purpose of this study is to evaluate the overall health status of MATC Truax's occupants in relation to the work environment. This study proposes to estimate the level of various illnesses and symptoms in the MATC Truax population. Observed levels of health status indicators will be compared with available community or baseline data to determine if observed levels of illness or symptoms exceed what would be expected in "normal" populations. This study is not designed to evaluate the reports of sarcoidosis because National Jewish Medical and Research Center, a leader in the field of sarcoidosis research, was approached to evaluate sarcoidosis specifically.

Methods

A random sample of full time staff and faculty at the MATC Truax campus were surveyed concerning the workplace environment and health status. Students and part-time faculty were excluded from the survey because building related effects are expected to be most evident in individuals present on campus for the majority of the workweek. Epi Info 6, an epidemiological software package produced by the Centers for Disease Control and Prevention, was used to calculate the sample size for the survey and generate a random list of numbers used to select the sample. The original sample included 265 randomly selected faculty and staff out of 611 that were listed in the most recent MATC Truax directory.

The MATC Truax survey questionnaire, based on the National Institute for Occupational Safety and Health's (NIOSH) Indoor Air Quality (IAQ) Survey questionnaire, was developed to answer some critical questions about health status and environmental condition/exposure. Parts of the NIOSH questionnaire were excluded in an attempt to shorten the questionnaire for this survey. This step was taken because the length of the NIOSH questionnaire may have dissuaded some survey recipients from completing the survey. Also, shortening the questionnaire helped to focus the responder's attention on the most pertinent questions.

The MATC IAQ Committee suggested that distribution of survey questionnaires by email would improve survey response because most faculty and staff rely on email heavily and pay close attention to the material received by email. The IAQ Committee also felt an electronic questionnaire would be efficient to distribute and easier and faster to complete and return. In early February 2000, questionnaires were distributed by e-mail to all but 10 of the randomly selected study participants. The remaining 10 surveys were delivered by hand to individuals without email addresses. The timing of the survey distribution was planned to occur a few months prior to spring break to assure that staffing would be at normal levels. Questionnaires were distributed with a cover letter explaining the purpose of the study and the need for participation from each study participant. Recipients were instructed on how to complete questionnaire electronically or by hand. Surveys were returned directly to Madison Department of Public Health by email, postal delivery or hand delivered. Individuals identified to be absent from campus at the beginning of the calendar year through the questionnaire distribution date were removed from the sample. For the next four weeks, a weekly email message was sent to all non-responders to remind them to complete and return the questionnaire. Approximately one month after the surveys were distributed, non-responders were contacted by telephone a minimum of two times over the next two to three weeks. During this time, MDPH staff responded to questions from MATC faculty and staff concerning privacy, survey purpose and other concerns. Upon receipt, questionnaires were reviewed for consistency. MDPH staff contacted responders by phone or email to verify answers that were unclear or inconsistent with the question.

Reported levels of clinically diagnosed illnesses were compared with levels of these illnesses reported to a local health care insurance provider, Wisconsin Physician's Service (WPS). This provider was selected because it is the primary provider for medical care insurance to MATC faculty and staff. WPS was able to provide the percentage of their clients diagnosed with each specific illness included in this study with the exception of mold allergy. This was determined by querying claims records from physician's and clinics for the appropriate diagnosis codes. In calculating these percentages, WPS queried only the calendar year 1999 and counted individuals only once for each illness. WPS percentages also excluded Medicare claims to reduce the bias an older population may have had on the percentages. WPS provided illness percentages for both the client group that includes MATC faculty and staff from all Madison area campuses and their entire Madison area client group. Observed levels of symptoms were compared with published data collected during surveys of self-identified "problem" buildings and "non-problem" buildings. Self-identified "problem" buildings (those considered to be a problem by owners or occupants of the building) were studied as part of the National Institute for Occupational Safety and Health program to evaluate the cause occupant reported problems. Data on "non-problem" buildings discussed in this report resulted from studies performed by Washington State Department of Labor and Industries and the US Environmental Protection Agency. Chi-square analysis was performed to determine if the illness or symptom levels in the MATC Truax population varied significantly from either the NIOSH or Washington referent populations. Pearson correlation coefficients were calculated for each variable pair to assess the relationship between health endpoints and workplace/environmental variables.

Results

Survey Response

Past health surveys of MATC Truax campus have failed because of the lack of response from those selected for the survey. In this survey, the attempts to improve the response rate were successful in garnering a 79% response from the individuals sampled. Eighteen individuals were removed from the original sample and were not included in determining the response rate because they no longer worked for MATC Truax or were away from the campus for a month or more prior to the survey. As determined by chi-square analysis, the responders did not differ significantly from the population of full-time staff and faculty at MATC Truax when compared across floors, room types, or position types. The high response rate and the relative similarity of the responder group to the Truax population indicate that this sample should be representative of the campus as a whole. It should be noted that the sample selected included staff from the main Truax building and the Administration building. The Administration building is separate from the main building and has separate air handling equipment. In reviewing the results, it was decided to leave the Administration building staff in the sample because it makes up only 6% of the total responder group and is unlikely to result in any significant bias. Survey participants from the Administration building were separated from other areas of the MATC Truax building when comparing health, workplace, and environmental factors between floors of the building.

Workplace Information

Several questions were asked in the survey to gather information on the type of work area in which the responders spend a majority of their time. These questions also attempted to gather information on the types of conditions that are present in these work areas. Most responders (43%) reported that their work area was an open space with partitions (n=189). Of these responders, 42% shared this space with 4 to 7 co-workers and 52% shared their space with 8 or more co-workers. When asked about the general cleanliness of their work area, respondents (n=189) were well distributed with most reporting either "Reasonably clean" (37%) or "Somewhat dusty or dirty" (42%). Table 1 provides the percent of responders sharing their work area with various common types of office equipment. When asked about recent changes to the area that responders occupy, few individuals reported that any changes in furnishings, wall coverings/coatings, or water damage had occurred over the last 3 months. Table 2 reports the response to the changes listed in the survey.

Some MATC Truax survey recipients had difficulty answering questions about their workplace. The difficulty resulted from the fact that many faculty and maintenance workers divide their time at MATC Truax among different work areas. Some employees also split time between MATC Truax and other sites off-campus. In some cases, responders spent 20% or less of their workweek at their primary workstation at MATC Truax.

Information about Health and Well-Being

A large percentage of the sample reported that they were non-smokers (77%, 151). Another 11% of responders considered themselves former smokers and 9% reported that they were current smokers. Three percent of the respondents did not answer the question. For this survey, a smoker was defined as a person that has smoked at least 100 times and is still smoking. This is much lower than what is observed in Dane County (20% current smokers) and Wisconsin (23% current smokers) (1).

Individuals selected for this survey were asked a basic set of questions to measure their current health status and gather information about past medically diagnosed illness. The survey asked recipients to report the most recent diagnosis of asthma, bronchitis, cardiovascular disease, diabetes, dust allergy, emphysema, eczema, fibromyalgia, influenza, migraine, mold allergy, sarcoidosis or sinus infection. Figure 1 identifies the percentage of individuals surveyed self-reporting diagnosed illnesses that developed or were identified in 1999. No cases of diabetes or fibromyalgia were reported in 1999; however, the 5-year average (1995 - 1999) for diabetes and fibromyalgia was calculated and included in Figure 1. Illnesses reported by MATC Truax survey participants without a diagnosis date (emphysema, 1 case; sarcoidosis, 1 case) were excluded from figure 1. As a comparison, the percentage of these illnesses reported in 1999 were compared to Wisconsin Physician's Service (WPS) data for MATC WPS Health Care Plan participants and Madison Area WPS Health Care Plan participants. The percentages of asthma, bronchitis, cardiovascular disease, diabetes, dust allergy, eczema and fibromyalgia reported by MATC Truax staff were not significantly higher than the percentages reported by WPS for all MATC Health Plan participants and Madison area participants. Chi-square analysis of asthma and cardiovascular disease levels identified a significant difference between the three populations. Chi-square analysis was not valid for diabetes or dust allergy because of the small number of reported cases in the MATC Truax sample. Percentages of influenza, migraine, and sinus infection reports in 1999 by MATC Truax survey participants were considerably higher than that reported for WPS client populations. Chi-square analysis found the study population and the two WPS client groups to be significantly different on this measure. Survey participants were given an opportunity to list additional diagnoses not listed. Illnesses reported by two to four individuals included hypertension, pneumonia, and seasonal allergies.

The third health status question asked survey participants the symptoms that they had experienced and the frequency of those symptoms over the last 4 weeks. For each symptom reported, responders were asked to report if that symptom went away after leaving the MATC Truax campus and how much time was required for this to occur. In the following tables or charts, abbreviations are provided for each symptom description. Table 3 lists each of these abbreviations and their corresponding symptom description. Figure 2 shows that many staff and faculty at MATC Truax reported frequent occurrences of the various symptoms listed. However for certain symptoms, many of those reporting a high frequency of the symptom also report that the symptom does not go away when they leave Truax. Figure 3 shows this is the case for "stuffy or runny nose", "sinus congestion", "difficulty remembering or concentrating" and "dry or irritated skin". Variation was observed among the floors/areas of the Truax campus, however, chi-square analysis among the three most populated floors/areas for each symptom did not find any significant differences among respondents on the floors/areas. Figure 4 compares the percentage of individuals reporting work-related symptoms with the average percentage of symptom prevalence from NIOSH HHE's (2) and from a study of non-problem buildings in Washington State (3). In this report, we are using NIOSH's definition of a "work-related" symptom. NIOSH defines a "work-related" symptom as one that has occurred at least one to three times per week over the last month and improves after leaving the workplace. No symptom levels were significantly greater than NIOSH averages.

Other indoor air quality studies have used symptom groups to evaluate the health status of study participants. One study was of non-problem state office buildings in Washington State (3). Definitions of these symptom groups are listed in table 4. Symptom group percentages for MATC were found to be less than the percentages published in the Washington State study (Figure 5). Neither symptom groups nor individual symptoms were correlated strongly either with workplace situations reported in section I of the questionnaire or perceived environmental conditions reported in section III of the questionnaire. Whether a person's primary workstation was located in an identified "hotspot" also was not correlated with individual symptoms or symptom groups.

Another method of evaluating reported symptoms is the development of a symptom index. An initial report from EPA's Building Assessment Survey and Evaluation (BASE) study developed a personal symptom index (PSI) which is the count of specific symptoms often related to sick building complaints (Table 4) (4). The average PSI for each building, termed the building symptom index (BSI), was then calculated to compare among buildings in the EPA BASE Study. Figure 6 shows that MATC Truax's BSI is similar to the average BSI calculate for 16 non-problem buildings included in EPA's BASE study.

Reported workplace conditions

The final section of the survey asked the responder to report the frequency of experiencing a variety of common indoor environmental quality conditions. The most frequently observed indoor environmental quality problems were "Too little air movement" and "Air too dry"(recorded in Figure 7 as "Stuffy" and "Dry", respectively). These perceived conditions were strongly correlated with each other (Pearson correlation coefficient = 0.64, p=0.0000). "Too little air movement" was also weakly correlated with "Temperature too hot" and "Unpleasant chemical odors" (Pearson correlation coefficient = 0.54, p=0.0000 and 0.42, p=0.0000 respectively). Other conditions reported by many responders as occurring frequently included "Temperature too hot", "Temperature too cold", "Unpleasant chemical odors", and "Other unpleasant odors".

Discussion

Evaluation of Reported Diagnosed Illness

WPS data on the illnesses discussed in this survey are the best referent data available because most of the illnesses included in this survey are not included in standardized disease surveillance tools used by state or local public health agencies. However, WPS client groups do not match MATC Truax staff and faculty perfectly. The MATC Truax population surveyed in this project included only working adults whereas both WPS populations include children, non-employed spouses, and retired employees. Populations with more young or elderly individuals may bias the number of illnesses reported by WPS higher because some illnesses are more likely to occur in young (asthma for example) or old (cardiovascular disease for example) individuals. The employment status of this referent population may also have an effect on the numbers of illness reported by WPS; however, the direction of this bias is difficult to predict. The WPS MATC client referent group more closely matches the study population than the Madison area WPS clients because those individuals holding MATC health care plan certificates are likely to have similar employment patterns to the faculty and staff at MATC Truax.

In addition to age and employment differences, the level of verification may differ between illness reported by survey participants and illness reported to WPS. Illnesses reported by MATC Truax staff and faculty have not been confirmed by a check of personal medical records. However, WPS data was extracted from reports submitted to WPS from physicians and medical offices for payment of services. Without examination of personal medical records, it is possible that some of the reports of illnesses among MATC staff were not based on diagnostic methods performed by a medical provider. This bias may tend to increase the reporting of these illnesses among MATC staff because a medical diagnosis is more stringent than self-diagnosis. It is not possible to judge at this time the extent to which self-diagnosis was reported in this situation. In general, we assume that survey respondents understood that this question was asking for physician-diagnosed illnesses and answered accordingly.

The differences in illness levels may not be completely explained by these differences between our referent and study populations. Influenza is known to result from infection of one or more influenza viruses (5). Also, individuals working in environments such as schools, where occupants meet and interact with many other people daily, are at higher risk of infection due to greater risk of influenza virus exposure. Studies have hypothesized that mechanical ventilation systems may increase the transmission of respiratory infections such as influenza and the common cold (6-8). However, this pathway of transmission remains unproven and very controversial in the scientific community. It should also be noted that other studies have found mechanical ventilation systems superior to natural ventilation based on health outcomes (9). More than any other illness discussed in this study, influenza has the greatest likelihood of being over-reported by the MATC Truax sample. The "flu season" is a regular occurrence in this region of the country; however, many people often confuse the symptoms typical of influenza with those of severe colds and gastrointestinal illness. Also, it is likely that the number of cases of influenza reported to WPS is conservative because persons with influenza may not seek medical attention and attempt to control the symptoms with over-the-counter medications. While influenza attack rates vary over time and among populations, studies have found influenza attack rates that range from 14.9% to 19.7% for adults studied (10). For these reasons, the increase in reports of influenza in the MATC Truax sample cannot reasonably be considered to be building-related.

The causes of migraine are not completely understood. Several studies conclude that genetics and environmental factors are among the main factors contributing to the experience of migraine (11-14). One study attributed as much as 50% of the variability between migraine sufferers and non-migraine sufferers to genetic factors (13). These studies are not clear about what environmental factors are most likely to lead to migraine. One study reported that environmental factors leading to migraine were dependent on the individual (12). Another study reported computer play, loud noise, and hot climate to be the most common environmental exposures leading to migraine in children (14). At least one medical text, that discusses migraine, lists odor and foods along with genetics and other stimuli, including stress that may result in migraine attacks (15). While strong environmental factors leading to migraine have not been well established, the MATC Truax sample reported much greater levels of migraine than either referent group. This difference is not well explained by age differences between the sample and referent groups because incidence of migraine is generally considered to be higher in children and become less frequent and severe with age (15). Migraine may be over-reported by self-diagnosis but it is also likely to be under-reported by physicians because individuals with mild cases of migraine may not seek medical attention. During the course of this study, several individuals outside of the sample reported that they experienced frequent migraine episodes while employed at MATC Truax but since leaving the facility, their migraine episodes had decreased or stopped completely. This anecdotal evidence combined with the increased reports from the MATC Truax sample provide reasonable cause to evaluate more closely the relationship between migraine and the MATC Truax campus.

Sinus infection was reported most frequently of the illnesses discussed in this survey. Sinus infections are known to be bacterial infections of the facial sinuses that result from blockage that prevents normal drainage (16). Sinus infections commonly occur secondary to respiratory infection or allergic rhinitis. They also may occur after exposure to chemical irritants that impair drainage of nasal secretions. For these reasons, sinus infection may be indirectly related to environmental exposures. While it could be suggested that an observed increase in influenza and possibly non-seasonal allergies would lead to an increase in sinus infection, this would not explain the magnitude of the increase in sinus infections observed in this study. Other respiratory disease or allergies not included in this survey may have an influence on the level of sinus infection observed. However, survey responders did not list additional respiratory illness other than a few cases of pneumonia and seasonal allergies. While illness verification may be another source of the variation in this sample, it is reasonable to assume that self-diagnosis of sinus infection would be relatively low because of the pain and discomfort associate with this illness. Also, it is reasonable to assume that most insured individuals with sinus infection would seek medical attention for the same reason.

Diagnosed illnesses found to be similar among all three groups included bronchitis, dust allergy, eczema and fibromyalgia. Smoking is reported to be the most common single factor in the development of bronchitis (17). However, smoking and bronchitis were not significantly correlated with one another in this study. There are many other factors related to bronchitis including environmental, occupational, infectious, and genetic variables. Eczema is a broad term that includes skin reactions to allergens and infectious agents and is likely to be affected by exposures outside of the occupational environment. Allergic rhinitis is the medical term for the group of eye and upper respiratory symptoms commonly experienced by individuals suffering from allergies to dust, mold, pollen, and other airborne particulates. While dust allergens are undoubtedly present at MATC Truax as they are in most indoor environments, these data do not identify an increase in dust allergy among MATC Truax staff over the general population. The number of individuals reporting mold allergies exceeded the number reporting dust allergies. However, this study cannot evaluate the observed level of mold allergies because WPS was unable to report the percentage of their clients diagnosed with mold allergies due to the lack of specific diagnosis codes for mold allergy. Allergic rhinitis is generally seasonal due to the seasonal nature of many allergens (18). However, perennial allergic rhinitis will develop, generally during adult life, when the allergen is always present such as with indoor dusts and molds. Reporting bias is a factor to consider when evaluating the observed level of dust allergies in the study group. Referent populations are known to include children that would have a greater risk of being diagnosed with dust allergies than an adult population.

Asthma, cardiovascular disease and diabetes were found to be lower then the referent populations in this study. Cardiovascular disease and diabetes are not considered to be strongly related to the occupational environment and was included in this study to address concerns of reporting bias. Reasons for the lower observed levels of cardiovascular disease and diabetes in the study population are unknown although the low smoking rates may play a role in the lower level of cardiovascular disease. Illness reports from the MATC Truax sample also identified lower levels of asthma in the sample than in the referent groups. It is possible that age differences between the sample and referent groups accounted for some of this variation (19). It is well known that asthma diagnosis is more prevalent in children then adults. If the WPS data were specific to adult onset asthma, this comparison would be more meaningful.

Evaluations of reported symptoms

At this time, no data exists on symptom prevalence for "clean" buildings. In this case, "clean" refers to a building that is free of all sources of indoor air pollutants. This makes it difficult to determine if the observed symptom levels are "normal". This difficulty continues as we attempt to compare observed levels of symptoms at MATC Truax with that observed in other studies. The comparisons made in this report are with NIOSH data for self-reported problem buildings, data from non-problem state office buildings in Washington, and EPA data from their Building Assessment Survey and Evaluation (BASE) study. A "non-problem" building is one in which sources of indoor air pollution may exist but air quality or health problems have not been reported to the extent that would label the building a "problem". It cannot be assumed that all occupants of non-problem buildings are free of building-related illness (3). Comparison with NIOSH data is problematic because symptom levels that are lower than the reported NIOSH levels do not necessarily indicate a normal level of the symptom (20). Also, NIOSH found that most symptoms from their study of self-identified problem buildings had similar prevalence rates to the four non-problem state office buildings studied in Washington (2). This led NIOSH to suggest that the baseline for health symptoms in office buildings is relatively high and that the determination that a building is a "problem" building may be related to factors other than symptom prevalence. In this study, we compared symptom prevalence from MATC Truax study participants with data from both of these studies. No symptom levels were found to significantly exceed that seen in the NIOSH study; however, several individual symptoms were similar to that observed in the self-reported problem buildings studied by NIOSH and the non-problem state office buildings in Washington. Comparison of the BSI for MATC Truax with the average BSI for non-problem buildings evaluated in EPA's BASE study supports the suggestion that a relatively high baseline of symptoms exist in office buildings including MATC Truax (4).

Correlation of health indicators and environmental conditions

The final step in this study was to determine if relationships exist between illness or symptoms and environmental or workplace factors. As stated above, individual symptoms, symptom groups or clinically diagnosed illnesses, including migraine, sinus infection, and influenza, were not strongly correlated with any of the personal, workplace, or environmental factors collected in the survey of MATC Truax. For this reason, further regression analysis was not performed on the MATC Truax data. Correlations between illness or symptom reports and the amount of time employed at MATC Truax or located in the current workstation were weak. However, the question of time at current work station or employed at MATC Truax did not answer the question of how much time the survey participants spent in the facility for a given work day or work week. This is an important aspect that should be included in any additional studies of health status.

Conclusions

In summary, observed levels of migraine, and sinus infection in MATC Truax full-time faculty and staff significantly exceeded that reported for local referent populations. Observed levels of influenza are also significantly greater than levels reported by WPS; but it is unlikely that this difference is related to environmental conditions at MATC Truax. Symptom reports, on the other hand, were not found to be different than those from non-problem buildings in the published literature and were not strongly correlated with any of the personal, workplace, or environmental factors collected in this study. These findings are not necessarily at conflict in this study. To start with, the time frame in which individuals were asked to report differed for illnesses and symptoms. Survey participants were asked about symptoms that occurred in the last month but were asked to report any diagnosed illnesses (the discussion above focused on diagnoses occurring in 1999). This time difference is unavoidable in studies of this type because symptom reports over long periods of time become suspect due to recall bias. At the same time, limiting the report of diagnosed illnesses to those occurring over the last month unnecessarily limits the ability of the study to evaluated physician diagnosed illness that is less susceptible to recall bias than symptom reports. Thus it is not reasonable to compare symptom and illnesses reports directly. Also, the discussion of reported symptoms focused on those considered to be "work-related". This is a valid attempt to control the fact that symptoms may be related to countless environmental factors that survey participants are exposed to outside of the workplace. However, NIOSH suggests that it is not necessarily appropriate to assume that a symptom related to a workplace exposure will improve after leaving the workplace (2). Some symptoms may persist longer than a few hours to a few days and leaving the workplace may not eliminate some workplace exposures.

While some illness may be elevated in the MATC Truax population, this data cannot conclude a cause and effect relationship between these illnesses and the MATC Truax environment. Further study is necessary before a cause and effect relationship can be determined. This study was not designed to observe if rooms or sections of the Truax campus have greater levels of illness or symptoms than others do. The lack of observed increases in illnesses or symptoms, other than sinus infection and migraine, throughout campus does not necessarily imply that specific rooms or sections within the main building of campus do not have environmental issues that need to be addressed. Environmental evaluations have been conducted by NIOSH and the National Jewish Medical and Research Center in rooms and sections of the main building considered to be "hotspots" based on the number of complaints received from those rooms. These reports should be used to address concerns in these specific areas of high concern.

Data from this study support the following recommendations:

1. MATC should pursue further epidemiological study to determine if there is a cause and effect relationship between the development of sinus infection or migraine and the indoor environment of MATC Truax.
2. Independent of further epidemiological study, MATC should continue to correct problems identified by their engineering consultants, National Jewish Medical and Research Center and NIOSH with the goal of correcting existing problems and preventing future problems.
3. Independent of further epidemiological study, MATC administrators, faculty, staff, and students should continue to openly and clearly communicate information on indoor air quality concerns, results of studies and analysis, and actions taken to correct and prevent exposure to poor indoor air quality.

References

1. Madison Department of Public Health. A Picture of Health, An assessment of the health of Madison and Dane County. 1995.
2. Malkin, R, T Wilcox, and WK Sieber. The NIOSH Indoor Environmental Evaluation Experience: Part Two. Symptoms Prevalence. Applied Occupational and Environmental Hygiene, 1996 11(6):540-545.
3. Nelson, NA, JD Kaufman, J Burt, and C Karr. Health symptoms and the work environment in four non-problem United States office buildings. Scand J Work Environ Health, 1995, 21:51-59.
4. Brightman, HS, SE Womble, EL Ronca, and JR Girman. Baseline information on indoor air quality in large buildings (BASE '95). Indoor Air '96: Proceedings of the Seventh International Conference on Indoor Air Quality and Climate, 1996, Nagoya, Japan.
5. Dolin, R. Influenza. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 814 -819.
6. Cunny, RJ, A Bialachowski, D Thronley, FM Smaill, and RA Pennie. An outbreak of influenza A in a neonatal intensive care unit. Infect Control Hosp Epidemiol, 2000 21(7):449-454.
7. Drinka PJ, P Krause, M Schilling, BA Miller, P Shult and S Gravenstein. Report of an outbreak: nursing home architecture and influenza-A attack rates. J Am Geriatr Soc, 1996 44(8):910-913.
8. Jaakkola JJ and OP Heinonen. Shared office space and the risk of the common cold. Eur J Epidemiol, 1995 11(2):213-216.
9. Niven RM, AM Fletcher, CA Pickering, EB Faragher, IN Potter, WB Booth, TJ Jones and PD Potter. Building sickness syndrome in healthy and unhealthy buildings: an epidemiological and environmental assessment with cluster analysis. Occup Environ Med, 2000 57(9):627-634.
10. . Glezen WP and RB Couch. Influenza Viruses. Viral Infections of Humans, Epidemiology and Control, 3^rd ed. Plenum Medical Book Company. pp. 419-449.
11. . Ulrich, V, M Gervil, KO Kyvik, J Olesen, and MB Russell. The significance of genetic and environmental factors for migraine with aura. A genetic epidemiologic study of Danish twins. Ugeskr Laeger, 2000, 162(2):167-170.
12. . Gervil, M, V Ulrich, J Kaprio, J Olesen, and MB Russell. The relative role of genetic and environmental factors in migraine without aura. Neurology, 1999, 53(5):995-999.
13. . Ziegler, DK, YM Hur, TJ Bouchard Jr, RS Hassanein and R Barter. Migraine in twins raised together and apart. Headache, 1998, 38(6):417-422.
14. . Bener, A, SA Uduman, EM Qassimi, G Khalaily, L Sztriha, H Kilpelainen and E Obineche. Genetic and environmental factors associated with migraine in schoolchildren. Headache, 2000 40(2):152-157.
15. . Raskin, NH. Headache. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 65-71.
16. . Lebovics, R and AS Baker. Infectious diseases of the upper respiratory tract. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 515 - 520.
17. . Ronald H and JR Ingram. Chronic Bronchitis, emphysema, and airways obstruction. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 1197-1206.
18. . Austen, KF. Diseases of immediate type hypersensitivity. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 1630 - 1638.
19. . McFadden, ER Jr. Asthma. Harrison's Principals of Internal Medicine, 13^th ed. McGraw-Hill, Inc. pp. 1167 - 1172.
20. .  Mendell, MJ, WK Sieber, MX Dong, R Malkin, and T Wilcox. Symptom Prevalence distributions in US Office buildings investigated by NIOSH for indoor environmental quality complaints. Indoor Air '96: Proceedings of the Seventh International Conference on Indoor Air Quality and Climate, 1996, Nagoya, Japan, 2:877-882.