Cure of a sick building: A case study

Environment International, Vol. 15, pp. 107-114, 1989 Printed in the U.S.A. All rights reserved.

0160-4120/89 $3.00 +.00 Copyright @1989 Pergamc~ Press plc

CURE OF A SICK BUILDING: A CASE STUDY Knud J. Helsing Department of Epidemiology,Johns Hopkins School of Hygiene and Public Health, Baltimore, MD 21742-2067, USA

Charles E. Billings Medical Department, MassachusettsInstitute of Technology,Cambridge, MA, USA

Jose Conde Washington County Health Department, Hagerstown, MD, USA

Ralph Giffin Washington County Board of Education, Hagerstown, MD, USA E1 87-343 (Received 20 October 1987; Accepted 27 February 1989) This case study illustrates the crisis atmosphere that can develop if occupants of a building get the feeling that their complaints are being ignored. Following several years of complaints about stale air and poor temperature control, allegations were made publicly that faculty and students were being poisoned by high levels of carbon monoxide (CO) and carbon dioxide (CO=) in a five-year-old school. As a result, a multidisciplinary study was initiated to assess: 1) available records of health and academic achievement of students; 2) actual levels of CO and other contaminants in the air; and 3) measurable ventilation characteristics of classrooms. No excess absenteeism, illnesses, or lack of academic achievement were found, nor were any significant levels of CO, HCHO, or CO2 found. However, there was an insufficient fresh air supply to some classrooms, and a large percentage of students exhibited classic symptoms of Sick Building Syndrome, i.e., headache, eye burning, and fatigue. Correcting the ventilation problems resulted in a reduction of symptoms to a level approximately equal to that of students in other schools in the county.

INTRODUCTION

However, not all of the above symptoms need be present to classify a building as sick. Rather, any cluster of several symptoms experienced by a substantial number of a building's occupants could be adequate cause for investigation. The mix of symptoms could vary, depending on the nature of the air pollutants contributing to the problem in any particular building. The Northern Middle School in Hagerstown, MD, seemed to meet the requirements of a sick building, and the investigation brought out general information that may be helpful in other instances.

According to Bahnfleth (1986), Carlton-Foss (1983), EURO Reports and Studies No. 78 (1983), and Hodgson et al. (1985), the Sick Building Syndrome is by no means a new phenomenon. In 1982, a World Health Organization (WHO) working group on Indoor Air Pollution listed eight symptoms characterizing the disease: 1. Irritation of eyes, nose, and throat 2. Dry mucous membranes and skin 3. Erythema 4. Mental fatigue, headache 5. Airway infections, cough 6. Hoarseness of voice, wheezing 7. Unspecified hypersensitivity reactions 8. Nausea, dizziness

BACKGROUND Northern Middle School is a modern, two-story, brickand-concrete building located in Hagerstown, MD. It was constructed in 1979 and occupied in January 107

108

1980 with the transfer of the entire student body from a nearby obsolete building. Northern Middle is designed to be energy efficient - it has unopenable windows and Variable Air Volume (VAV) ventilation supply to individual rooms and spaces. The main building-ventilation supply units are located in two penthouses on the roof, with flesh air intakes on the rear wall of each penthouse. Outside air is mixed with building return air in the penthouse plenums; filtered; pressurized by supply fans with variable air inlet dampers; and either heated or cooled, depending on season, to about 50°F (100C). Main supply ducts distribute conditioned air to each of the 19 individual classrooms, 10 large open areas, numerous administrative offices, and other building facilities. Ventilation air enters each room through slot diffusers in the dropped ceiling and exits to the space above the dropped ceiling (and from there to the penthouses) through slots in the recessed fluorescent light fixtures. Water heated by an oil-fired boiler circulates through pipes at the building perimeter to supply heat in the winter; temperature control is maintained by using a VAV box controlled by the thermostat in each room to vary the amount of 50°F (10°C) air entering each room. The gymnasium and locker rooms have independent air-handling systems, the latter using 100% fresh air with no recirculation. Almost immediately after the school was occupied there were complaints by the faculty of inadequate air circulation, poor temperature control, odors, headaches, dizziness, etc. In 1982, an engineering consultant was called in and recommended a number of changes in the operation of the heating and cooling system and some modification in the design of the exhaust and outside air intake mechanisms. Among the changes made was the addition of about 12 feet to the height of the smokestack serving the oil-fired furnace. (The consultant had found that under some wind conditions the smoke from the original stack could be carried directly into the fresh air inlet of the air handling system serving the western half of the school.) Complaints continued, however, and in December 1984 a written report by a faculty member alleged that elevated levels of CO and CO 2 in the building were endangering the health of faculty and students. Based on this report, the parent-teacher organization and the citizens' advisory board demanded that either immediate action be taken to correct the problems or the school be closed. In this crisis atmosphere, the Board of Education requested the help of the Department of Environmental Health of the Johns Hopkins School of Hygiene and Public Health. A multidisciplinary investigation

K.J. Helsing

was initiated to determine the magnitude of the problem and to recommend corrective measures where needed. INVESTIGATION AND FINDINGS

Several approaches to the determination of the magnitude of the problem were adopted concurrently: A. Immediate investigation of existing records of health-related behavior of students and teachers to determine if evacuation of the school would be justified. B. Physical examinations of teachers and staff. C. Survey of problem areas and measurements of air quality and quantity. D. Epidemiologic assessment of symptoms among students.

Investigation of Existing Records Fortunately, a number of records that allowed for an immediate gross assessment of the health status of the students and faculty were available. Student absenteeism records for the five years before and after the move into the new school showed no increase after the move, and the rate of student absenteeism (Table 1) compared favorably with that of other schools in the county. Similarly, faculty absenteeism for illness was no higher than that in other schools, and visits to the "Health Room," where students go when they do not feel well, were no more frequent in Northern than in the one other county middle school with comparable available data. These data were somewhat reassuring in that if there had been any major deleterious effects on the health of the faculty or students, they would have been expected to manifest themselves in greater absenteeism or greater frequency of visits to the health rooms. Existing records were also utilized to investigate the allegation that the learning ability of the children had been affected by the school environment. Two types of tests were on record for all the middle schools in the county - - IQ tests and academic achievement tests. A careful analysis of the mean IQs of the students showed Northern Middle to be consistently at the top or second every year for the five years before and after moving into the new building. Similarly, the academic achievement of the student body showed no deficit among students of Northern Middle School; in fact, as seen in Table 2, they had a lower percentage of underachievers and a higher percentage of overachievers in one or more fields than students of comparable schools. At this point, the Washington County Health Department concluded that there was no evidence to warrant the closing of the school for health reasons.

Cure of a sick building

109

Table 1. Mean four-year percentage student absenteeism for five middle schools in Washington County, MD, during school years 1975-76 to1978-79and1980-81to1983-84. Time period School Northern

1975-79

1980-84

6.4

5.9

Change -0.5

Boonsboro

6.2

6.2

-

E. Russell Hicks

6.6

6.3

-0.3

Smithsburg

4.7

4.8

+0.1

Springfield

5.2

4.9

-0.3

Table 2. Net percentage of students in four middle schools who scored significantly higher or lower than their anticipated achievement as projected from each student's individual School Ability Index (IQ). Achievement areas scored* were Total Reading, Total Language, and Total Math. Northern Number of eighth graders tested Mean IQ of those tested

Middle School Clear Spring Smithsburg

Sprinqfield

208 105

129 104

206 106

247 103

11.5% 2.9% 0.0%

15.5% 3.9% 0.8%

14.1% 2.9% 0.5%

11.8% 4.1% 0.4%

Total percent underachieved in one or more areas

14.4%

20.2%

17.5%

16.3%

Net number of areas overachieved: 1 2 3

17.8% 5.8% 0.5%

16.3% 3.1% 1.6%

16.5% 2.9% 0.5%

17.5% 3.7% 0.4%

Total percent overachieved in one or more areas

24.1%

20.9%

19.9%

21.6%

Net number of areas underachieved: 1 2

3

*Scoring was done by Education Computer Software, Inc.

Physical Examinations Physical examinations of faculty and staff showed no evidence of common infection or allergy, and carboxyhemoglobin levels reflected only the normal smoker/nonsmoker status of the subjects. Pulmonary function studies were normal. Subjects with mouth and throat complaints evidenced mouth and throat irritation, and those with conjunctival complaints showed evidence of suffusion and, occasionally, mild

edema of the conjunctivae. Five of every six faculty members reported headaches, which tended to occur at work late in the day and which were associated with another highly prevalent symptom, fatigue. Occasional dizziness was also reported by a substantial number of the faculty and was usually associated with fatigue. Although the prevalence of some symptoms was high, none of the symptoms were serious enough to pose a hazard to health.

110

Air Monitoring While the reviews of existing records were going on, fixed, nondispersive, infrared analyzers were being installed in order to monitor both the CO levels of the air entering the air intakes of the ventilating system and the CO levels in two relatively open areas of the school (one on each floor). An instrument was also installed to monitor SO 2 level of the fresh air intake to the west penthouse, which was nearest the chimney. The CO was monitored because of the insistence by faculty and parents that it was the source of the problem; the SO 2 sensor in the intake air duct was added as a verifier of combustion by-products, if present. Thus, a rise in CO at the fresh air intake, if accompanied by a rise in SO 2, would indicate that combustion products might be present and a recording wind vane would point to the origin. At the request of teachers, six portable CO monitors were also obtained and placed in various rooms of the school. These proved to be less than satisfactory, however, because even insignificant changes in indicated CO levels within the limits of the accuracy of those instruments, such as from 1 pL/L to 3 [tL/L, tended to cause unnecessary concern. Furthermore, the instruments were found to be sensitive to electromagnetic radiation and to the presence of some aromatic hydrocarbons, such as eau de cologne. The only actual rise in the CO level occurred on the day when a trucker delivering food to the kitchen left his engine running and the door o f the loading dock open, causing a rise in the CO level in the cafeteria to about 35 gL/L for a few minutes. For a two-month monitoring period during the winter, when the infrared analyzers measured air entering the building, CO levels never exceeded 4 I.tL/L and were usually less than 1 ~L/L. Thus, CO was ruled out as a possible cause of the complaints. Similarly, SO 2 levels averaged from 0.009 to 0.028 gL/L, indicating no problem from smokestack emissions. Tests in several areas of the school showed formaldehyde levels ranging from 0.005 gL/L to a maximum of 0.076 gL/L, even with the ventilation system shut down for a weekend. These values are well below the 0.1 gL/L limit recommended by ASHRAE Standard 62-1981 (ASHRAE 1981). A smoke test of the sewer system disclosed three major potential sources of sewer gas into the building: there were two 8 cm openings to untrapped sewer vents in the cafeteria kitchen and one large gap in the vent piping above the girls' locker room - - a room that had been a major source of complaints. An investigation of the cleaning materials used by the custodial staff found that all materials on hand

K.J. Helsin8

were supplied by the Board of Education and were used in all schools in the county. "Grab sample" tests for organic vapors showed almost all detectable organics to be below 1 nL/L; the only exceptions were toluene, 2 nL/L, tetrachloroethylene (perchlorethylene), 1 nL/L, and his (2-ethyl hexyl) phthalate, 20 nL/L, all well below federal and state workplace standards. N u m e r o u s tests w e r e made with a p o r t a b l e CO 2 m e t e r in c l a s s r o o m s , b u t no r e a d i n g s o v e r 2.5 mL/L CO 2 were obtained. Thus, CO 2 was ruled out as a potential hazard. Airflow tests were conducted in several classrooms that had recorded numerous complaints. Total airflow in these rooms was marginal, with calculated fresh air supply under some conditions falling far below the 2.4 L/s per occupant recommended for school classrooms. In addition, dilution studies using sulfur hexafluoride as a tracer gas found the mixing factor K, defined as the ratio of total airflow to effective airflow, was as high as 4 in one of the classrooms with the lowest total airflow. With as little as 1/4 of the air supplied to the room effectively mixing, even less of the fresh air supply was reaching students seated at their desks. The investigation of fresh air inlets was also productive. A center portion of the second floor is recessed from the front of the school, forming a natural U-shaped pocket in which exhaust fumes from the school buses tended to collect when the wind was right. The locker rooms were served by an air-handling system that had air intake and outlet openings on one wall of this pocket, so even when bus exhaust fumes were not entering that system, the discharged air would tend to recirculate. Another undesirable situation was found at the west penthouse, where the air exhaust from the school's bathrooms was immediately adjacent to grilles opening into the penthouse that acts as a plenum for the second air-handling system. Any time a prevailing west wind was blowing, the bathroom exhausts could go right back into the penthouse to mix with recirculated air.

Epidemiologic Survey In May 1985 a questionnaire was mailed to the parents of eighth-grade students in Northern and two other comparable-sized middle schools in Washington County. Springfield Middle School is also a modern, energy- efficient school with windows that do not open, but it uses electric heat. E. Russell Hicks is another middle school with an oil-fired furnace, but which uses radiators for heating and windows that open for cooling in warm weather. The question-

Cure of a sick building

I 11

naire asked, among other things, how often a child had experienced any of 17 listed symptoms, 15 of which might conceivably be environmentally caused. Participation was 75% from Northern, 70% from Springfield, and 65% from E. Russell Hicks. Table 3 shows the percentage of symptoms reported to occur more than twice a month by parents of eighth-grade students in the three schools. Headache, overtiredness, and eye irritation or burning were reported significantly more often from Northern than from the other two schools. The percentage reporting sore throats or stomachaches was significantly greater in Northern than Springfield, but not E. Russell Hicks.

ACTION TAKEN

In order to correct the deficiencies found, a number of changes were made. The openings to the sewer line were, of course, sealed the same day they were found. Other changes involving some engineering design and construction were made during the school's summer vacation in 1985. Essentially, the changes were:

1. The supply fan speed in both main air handling units was increased in order to increase total airflow capability. 2. In six of the peripheral classrooms that had been the focus of most faculty complaints, larger Variable Air Volume terminal units were installed. Five of the replaced units were then installed in five other classrooms, replacing even smaller units. As a result, 11 classrooms received larger VAV terminals, thus permitting greater airflow than before. 3. Variable Air Volume control boxes for the large open interior spaces were modified so they could not open wide enough to take all the available air before it could reach the perimeter classrooms. 4. The supply air temperature was reset to 50°F (10°C) from the 58°F (14°C) setting that had been adopted as an energy saving measure. 5. Air intake to the locker rooms was changed so that air was obtained from the gymnasium supply air duct instead of the pocket of the building which tended to trap air exhaust fumes and school bus fumes. 6. A duct was built that carried bathroom exhaust air above the penthouse roof in order to avoid recirculation of that air.

Table 3. Percentage of eighth grade students in three middle schools reported by their parents to exhibit specified symptoms more than twice a month.

Washington County, MD May 1984 Symptom

Asthma Stomachache or cramps x Sore throat x Earache Hurt all over Fainting Stiff neck or back Headache 2 Vomit L o s s of a p p e t i t e Overtiredness 2 Eye irritation or burning 2 P a i n in c h e s t Dizzy spells B l o o d i n u r i n e o r B.M. Nervousness Nosebleed

Northern

Springfield

E. R u s s e l l

156

165

151

1.3 7.1 4.5 2.6 1.3 3.2 25.0 1.3 5.8 22.4

1.8 0.6 0.6 14.5 0.6 2.4 6.1

1.3 6.0 3.3 2.0 1.3 2.6 13.2 2.0 2.6 11.3

12.2 0.6 0.6 12.2 3.2

1.8 1.2 6.7 1.2

1.3 2.7 1.3 6.0 2.0

Hicks

tDifference between Northern and only one other school (Springfield) significant at p < 0 . 0 5 . 2Difference between Northern and both other schools significant at p < 0.05.

K . J . Helsing

112

7. In the course of the other improvements, workers discovered and repaired two broken supply air duct lines serving office and music areas. RESULTS

A tracer gas retest of Room 206, which previously had tested as the worst of the peripheral classrooms, with only 69 L/s effective ventilation at a 125 cm height from the floor, now tested 180 L/s under the same conditions. This volume, with a minimum of 33 % fresh air and a normal occupancy of 24 students, provides at least 2.4 L/s fresh air per student, the ASHRAE recommended minimum. Tests of airflow in other rooms showed that corrective measures were effective in increasing airflow to rooms needing it. One dramatic result of the action taken was the immediate drop in faculty complaints as the 1985-86 school year began. This trend continued all year. There were only 21 Energy Reports (written complaints concerning air temperature or other conditions) for the entire 1985-86 school year; these involved mostly complaints of temperatures being too low or too high. This compares to a total of 91 Energy Reports during the previous school year, most of which were complaints of stuffiness, headache, or eye irritation. Many or these reports contain multiple signatures, indicating complaints about entire areas of the school. A final epidemiological inquiry was made in May 1986. The same questionnaire that had been mailed to parents of eighth-grade students in Northern, Springfield and E. Russell Hicks Middle Schools in 1985 was mailed to parents of eighth-grade students of the same schools. Table 4 shows the 1986 response rate and compares it with that of 1985. The 1986 response rate was approximately the same for the three schools, whereas in 1985 the Northern response rate was five percentage points higher than Springfield and 10

points higher than E. Russell Hicks. It is possible that the difference in 1985 reflected increased concern among parents of Northern students caused by the considerable amount of newspaper publicity that had been given to the accusations made at that time. Table 5 shows the 1985 and 1986 percentages of Northern Middle School students reported to have experienced the specified symptoms more often than twice a month. It is obvious that all the symptoms that were significantly more prevalent in Northern than in one or both of the other middle schools in 1985 had dropped considerably in 1986. Table 6 shows the percentage of eighth-grade students in each school who in 1986 were reported to exhibit the listed symptoms more often than twice a month. There were no statistically significant differences between schools in any of the symptoms, using Yates X 2 and p < 0.05 as the criterion of significance. DISCUSSION

Several points learned in this investigation seem to be applicable to other situations: 1. If c o m p l a i n t s go u n h e e d e d , or if o c c u p a n t s o f a b u i l d i n g are not c o n v i n c e d that c o r r e c t i v e a c t i o n is b e i n g taken, the situation is bound to become worse. The same room conditions that caused drowsiness last year may now cause fatigue and headaches, the latter possibly a result of the frustration and anger over the situation. In the case of this school's problems, the accusations concerning damage to the health of the students and demands that the school be closed surfaced only after several years of milder complaints; although some improvements had been made to the ventilation system, the teachers were unconvinced that they were adequate. 2. It is surprising how much helpful information may be immediately available. Absenteeism for reasons

Table 4. Response rates to mailed questionnaires sent to parents of eighth grade students in three middle schools in Washington County, MD in May 1985 and May 1986.

Northern

Springfield

E. Russell Hicks

1985

1986

1985

1986

Enrollment

207

215

240

193

266

227

Number returned

156

141

167

132

173

143"

75.4

65.6

69.6

68.4

65.0

Percent returned

1985

*Does not include one blank questionnaire returned but considered a non-response.

1986

63.0

Cure of a sick building

113

Table 5. Percentage of eighth grade students in Northern Middle School reported by their parents to exhibit specified symptoms more than twice a month in 1985 and 1986.

Washington County, MD 1985

1986

156

141

1.3 7.1 4.5 2.6 1.3 3.2 25.0 1.3 5.8 22.4 12.2 0.6 0.6

0.7 4.3 2.8 1.4 1.4 0.7 2.8 12.8 0.7 2.8 14.2 5.0 1.4 1.4

12.2 3.2

6.4 0.7

Total n u m b e r reporting Asthma S t o m a c h a c h e or cramps x Sore t h r o a t ~ Earache Hurt all over Fainting Stiff n e c k or b a c k Headache 2 Vomit Loss of appetite Overtiredness 2 Eye i r r i t a t i o n or burning 2 Pain in chest D i z z y spells Blood in urine or bowel movement Nervousness Nosebleed

lDifference between Northern and Springfield Middle School was signifi: cant at p < 0.05 in 1985 2Difference between Northern and both Springfield and E. Russell Hicks school was significant at p < 0.05 in 1985.

of illness, for example, is a useful indicator of the severity of any problem that exists, and any business occupying a building will likely have records on the subject. Large organization may also have a nurse on staff who maintains more detailed records than those for just absenteeism. In addition, in newer buildings

with more sophisticated control systems, it is possible that records of periodic temperature and humidity readings in various parts of the building are available for review. In the absence of already available records, a pretested questionnaire may pinpoint particular symp-

Table 6. Percentage of eighth grade students in three schools reported by their parents to exhibit specified symptoms more than twice a month.

Washington County, MD May 1986

Asthma Stomachache or cramps Sore throat Earache Hurt all over Fainting Stiff neck or back Headache Vomit Loss of appetite Overtiredness Eye irritation or b u r n i n g Pain in chest Dizzy Blood in urine or bowel m o v e m e n t Nervousness Nosebleed

Northern

Springfield

141

132

142

0.7 4.3 2.8

O. 8 3.0 3.0

5.6 4.2 1.4 O.7 2.8 9.9 0.7 2.8 7.7 4.2 2.8 7. O -

1.4

-

1.4 O. 7 2.8 12.8 O. 7 2.8 14.2 5. O 1.4 1.4 6.4 O. 7

2.3 6.1 0.8 O.8 8.3 2.3 O.8 1.5 3. O O. 8

E. R u s s e l l

Hicks

114

toms or problem areas in a building and thus provide clues to the source of a problem. It is important that a questionnaire include some symptoms that are unlikely to be caused by the building environment, so as to detect possible hypochondriacs or individuals so emotionally upset over a situation that they will check any and all symptoms on a list. If at all possible, the questionnaire should also be presented to a comparable group somewhere else in order to avoid ascribing undue importance to a particular symptom like headaches. 3. Look for the obvious and easily located sources of contaminants first. In the present study it was surprising to find three large untrapped openings to the sewer line that apparently had existed ever since the building was first occupied. Finding exhaust air recirculating into the building seems to be less rare than one would hope, according to a study by Ferahian (1985); the present study is one more example supporting this conclusion. It took only an hour or so to determine that, at least in this building, fluorescent lighting fixtures, as studied by Terris e t a l . (1984), apparently were not contributing to the problem nor was flooding, as seen in a case study by Hodgson et al. (1985). Other localized and readily identified possible sources of indoor air problems that have been reported are electrostatic copying machines and carbonless-cepy forms in small, poorly ventilated rooms. Formaldehyde is another indoor-source contaminant that is readily measured if present in amounts likely to cause symptoms. 4. The initiatives discussed are not proposed as substitutes for a good survey of airflows and ventilation efficiency. They may, however, serve to focus part of an investigation on areas most likely to be deficient. In a study by A S H R A E (1981), it seems to be very important to ensure that the ventilating system be adjusted so that at no time is fresh air delivered to any occupied spacesbelow theASHRAE-rccommendcd amounts for the type of occupancy. 5. The psychological aspects of any occurrence of sick building syndrome must not be ignored. What may at firstglance appear to be frivolous complaints will often be found to give a clue to a problem area. As deficiencies are found and corrective measures undertaken, occupants of a building should be kept

K.J. Helsing

fully informed; complainers may be the first to acknowledge an improvement. 6. While room thermostats in this school remained nonadjustable, it should be recognized that, as studied by R y d e t al. (1985), there is now a trend among architects to give occupants some control over their individual work areas. It is likely that this will result in less dissatisfaction among occupants, although there may be some loss in energy efficiency. A study of this design change might be very productive. In the case of Northern Middle School, the fact that the school is located in a small city with relatively unpolluted air simplified corrective measures somewhat; when additional outside air was introduced, it was clean air. One minor exception must be noted - - one house across the street from the school used a wood stove for heating, and if the wind was just right, the smell of wood burning could be detected in the school. Until the source of that odor was explained to the teachers, there were some complaints. There is no doubt that a building in a densely populated area with heavy vehicular traffic can pose quite a difficult problem, but if the experience with Northern Middle School is any criterion, part of the problem may be of simple and readily corrected origin. REFERENCES A S H R A E . Standards for Natural and Mechanical Ventilation, A S H R A E Standard 62, American Society of Heating, Refrigerating and Air Conditioning Engineers, Atlanta, GA; 1981. Bahnfleth, D.R. Indoor air quality:problem of the 80's, A S H R A E J. 28, 30-31; 1986. Carlton-Foes, J.A. The tight building syndrome, A S H R A E J. 25, 38-41; 1983. Ferahian, R.H. Indoor air pollution caused by short-circuiting of fresh air intakes with exhausts of building. Proceedings of the CLIMA 2000 World Congress on Heating, Ventilating and Air Conditioning, Vol. 4, 307-312; 1985. Hodgson, M.J., Morey, ER., Attfield, M., Sorenson, W., Fink, J.N., Rhodes, W. W., et al. Pulmonary disease associated with cafeteria flooding, Arch. Environ. Health 40, 96-101; 1985. Ryd, H., Angelov, V., and Colvin, R. Climate and pedagogics. Proceedings of the CLIMA 2000 World Congress on Heating, Ventilating and Air Conditioning, VVS Kongres - VVS Moose, Charlottenlund, Denmark. Vol. 4, 313-320; 1985. Terris, D.R., Field, L., and Brumback, C.L, Outbreak of iUnes s due to volatilized asphalt coming from a malfunctioning fluorescent lighting fixture, Am. J. Pub. Health 74, 614-615; 1984. World Health Organization. EURO reports and studies No. 78; Report on a WHO working group, Indoor air pollutants: exposure and health effects, Copenhagen; 1983.