..

t1:·
r -.";

.... ., . _,.
........ •J,,L..'•··

SICK BUILDING SYNDROME: REVIEW AND

ABSTRACT TABLE 1
Control and abatement of indoor air quality (IAQ) Common Features of Symptoms Reported in Cases of
problems are dependent upon reliable investigation and Sick Building Syndrome (WHO 1983).
diagnosis. Sick building syndrome (SBS), building- Eye, Nose and Throat Irritation
related illness (BR!), and other health and comfort Sensation of Dry Mucous Membranes and Skin
problems are selectively reviewed and discussed. Psy- Erythema
chological and social as well as physical, chemical, and Mental Fatigue
Headaches, High Frequency of Airway Infections and Cough
biological factors that affect occupant physiological and Hoarseness, Wheezing, Itching and Unspecific Hypersensitivity
health responses are identified. Nausea, Dizziness
Confusion exists regarding definitions and attributes
of problem buildings. Timely, comprehensive, system-
atic investigations are rare, expensive, and difficult.
Systematic and other biases result in inadequate inves-
tigations and incomplete or incorrect diagnoses. Build-
ing ecology and building diagnostics are described as a TABLE 2
comprehensive framework for understanding and inves- Melhave's classification scheme for symptoms related
tigating indoor air quality problems. to sick building syndrome and examples of each
Hypothesized causes of SBS are identified based on (Molhave 1987).
published SBS and BR! investigation reports and review 1. Sensoric irritation in eye, nose or throat
articles. Methods to control SBS, BR!, and other build- dryness
ing-associated illnesses are presented and discussed. Pre- stinging, smarting, irritating sensation
hoarseness, changed voice
ventive measures to control IAQ-related health and 2. Skin irritation
comfort problems and recommendations for further re- reddening of skin
search are given. stinging, smarting, itching sensation
dry skin
INTRODUCTION 3. Neurotoxic symptoms
mental fatigue
Sick building syndrome (SBS) may affect as many as 20% reduced memory
of the office workers in the United States. In a survey of U.S. lethargy, drowsiness J
;
office workers, symptoms associated by respondents with reduced power of concentration
poor air quality included a tired, sleepy feeling (56%); a reduced memory
headache
congested nose (45 % ) ; eye irritations (41 % ) ; difficulty in dizziness, intoxication
breathing (40% ); and headaches (39%) (Woods 1987). Ef- nausea
forts to control and abate the causes of SBS in buildings are tiredness
potentially important to the economy and to public health. 4. Unspecific hyperreactions
running nose and eyes
The present paper is a review of selected published reports asthma-like symptoms in non asthmatic persons
related to SBS and a discussion of some of the problems respiratory sounds
limiting our understanding of it. 5. Odor and taste complaints
The complexity of modern buildings presents significant changed sensitivity
unmet challenges to designers, operators, and investigators. unpleasant odor or taste
Problems other than air quality can cause or exacerbate the
symptoms of sick building syndrome. Psychological and social
as well as physical and biological factors interact to create
occupant physiological and health responses to building en-
vironments. Yet detailed, comprehensive investigations of TABLE 3
building-associated outbreaks are infrequent due to the re-
WHO classification scheme for symptoms found in
sources and personnel required to conduct them. sick buildings (WHO 1986).
Control and abatement of SBS is dependent upon knowl-
1. Sensory irritation of skin and upper airways, along with headache
edge developed through reliable investigations and diagnoses.
and abnormal taste
Understanding the potential causes of SBS is essential to such 2. Odor
investigations and diagnoses. No clear understanding of SBS 3. General symptoms such as fatigue, dizziness and nausea
and no consistent definition of SBS is used. Authorities in 4. Lower airway and gastrointestinal symptoms (*)
the field use differing definitions or confusing terms which
impede progress in understanding the phenomenon. In fact, (*)Not generally found in sick building syndrome

263
the most widely accepted definition of SBS requires the ab- tamination by the causative agent and the illness is termed
sence of identified causes, but even those who present this building-related illness (see discussion below). But confusion
definition fail to use it consistently. exists because many investigators still apply the terin SBS to
Knowledge and understanding of SBS is obtained cases where symptom etiology is clearly identified.
through four primary means: (1) investigations of problem or
complaint buildings, with or without non-complaint control Elevated Rate of Symptom Reports
buildings; (2) multiple building studies, which may or may Some definitions of SBS require the demonstrated pres-
not include complaint buildings; (3) controlled experiments ence of excess reported symptoms in the complaint building
in buildings or laboratories where environmental factors are compared with a control building or some other comparable
manipulated and the responses of occupants are surveyed; baseline. This involves surveys of building occupants, usually
and ( 4) literature reviews where data or findings from various using epidemiologic techniques. However, the measurements
investigations or studies or both are collected and analyzed. made in most reported studies are insufficient to enable iden-
The investigations are usually commissioned by building own- tification of the pollutant concentrations and their associa-
ers, operators, or occupants while the studies and research tions with symptoms. There are usually too few measurements
are usually funded by public or private research or by gov- of environmental factors. Large occupant populations are
ernmental agencies. Reviews (including the present paper) studied through questionnaire surveys, either self-adminis-
are usually initiated by interested authors. tered or administered by the investigators. The large expense
of comprehensive pollutant measurements in each distinct
DEFINITIONS OF SBS AND OTHER BUILDING- environmental niche within a study building results in few
ASSOCIATED ILLNESSES samples of few pollutants in nearly all investigations. Large
spatial and temporal variability of indoor air pollutant con-
SBS is variously defined by its symptoms, by its hypoth- centrations within a single building or even spaces within the
esized causes, or by the demonstration of a statistically valid building (for C02, RSP, and VOC, for example) reduce the
association of SBS symptoms with a particular building. likelihood that one or a small number of measurements will
Symptoms provide adequate information to identify associations be-
tween exposures and symptoms. The "limitations of inves-
SBS is frequently defined by the occurrence of reported tigations" are explored later in this paper.
symptoms from a group of symptoms listed in several au- Under the elevated symptom prevalence definition, one
thoritative publications. Most of these definitions declare that or even a small percentage of a building's occupants cannot
the symptoms abate upon leaving the building and worsen suffer from sick building syndrome regardless of the symp-
upon re-entry (NAS 1981; WH·o 1983; Stolwijk 1984; Fanger toms or the building-relatedness of their onset and recurr-
1987; Finnegan et al. 1984; M0lhave 1987; Woods 1987). A ence.
widely cited World Health Organization report (WHO 1983)
lists a broad spectrum of symptoms reported primarily in
Scandinavia and the United States. Those symptoms have Sources of Confusion About Terminology
many features in common and are listed in Table 1. There are two sources for much of the confusion. One
M0lhave (1987) has proposed a classification scheme for involves definitional differences; the other involves incon-
the major symptoms of SBS (Table 2). A World Health Or- sistent use of terms. Where the etiology of symptoms and
ganization Working Group on Indoor Air Quality Research complaints is identified, most investigators do not label the
(WHO 1986) has proposed a classification scheme for "sick" symptoms or complaints SBS or the building "sick" (WHO
buildings (Table 3). These two authorities have presented 1983; M0lhave 1987; Woods 1987). Rather, they define the
conflicting schemes. Their lists are inconsistent, and Molhave symptoms as manifestations of illness and classify it as build-
has classified SBS symptoms while the WHO has classified ing-related illness (BRI) to distinguish it from SBS. BRI
"symptoms found in sick buildings." The WHO list includes includes such medical conditions as hypersensitivity pneu-
symptoms which it states are not found in sick building syn- monitis, Pontiac fever, and allergic dermatitis (Hodgson 1986;
drome, so a sick building is not equated exactly but overlaps Woods 1987). Building-related illnesses are attributed to a
with one in which sick building syndrome occurs. broad range of pollutants including infectious microorga-
nisms, allergens, chemicals, moisture, temperature, noise, vi-
Causal Factors bration, and poor illumination.
Commonly, several broad classes of factors are consid-
ered potentially related to an elevated incidence of reported Building-Associated Epidemics
symptoms when a problem is termed SBS (Skov and Valbjom Kreiss and Hodgson have used the term "building-as-
1987; M0lhave 1987). These factors include chemical sociated epidemics" to include both SBS and BRI (Kreiss and
(M0lhave et al. 1984), physical (Alsbirk 1983), biological Hodgson 1984).
(Morey 1984), and psychosocial factors (Colligan 1981; Alex- Other authors have used BRI as the all-inclusive term
ander and Fedoruk 1986). covering two categories of episodes: "those characterized by
Many authors state that most investigations of SBS have a generally uniform clinical picture for which a specific etiol-
not resulted in definitive identification of causal factors. In ogy can often be identified, and those in which affected work-
fact, most definitions of SBS require that the reported symp- ers report nonspecific symptoms temporally related to work"
toms not be associated with specific environmental or other (Samet et al. 1988).
causal agents. Other definitions of SBS expressly require that While asserting that SBS does not involve identified etiol-
the causal agent(s) not be clearly identified or demonstrated ogy, some authors still label cases as SBS which include bac-
by the investigation. Generally, where a causal agent is iden- terial diseases (such as Legionnaire's disease and Pontiac
tified, the symptoms are no longer considered "sick building fever), thermal discomfort, and irritation caused by chemi-
syndrome"; rather, the building problem is specified as con- cals, and they call such buildings examples of "sick" buildin~

264
 ..
(WHO 1986; Berglund and Lindvall 1987). Thus, confusion TABLE 4
is caused by the overlapping but different use of the terms
"sick building" and "SBS." Proposed classification scheme for occupant
condition in problem buildings (after Woods 1988).
Some authorities argue that buildings cannot be sick,"
they may only be contaminated. The term " sick building syn- 1. Unresolved.
Symptoms reported or complaint rates deemed unacceptable
drome" may just inappropriately medicalize an engineering, by owners, operators or occupants but do not meet standard
architectural, or maintenance problem (Hodgson 1989). statistical tests to confirm their association with occupancy of
the building. This can occur where contamination or complaints
Sick Buildings are limited to a small area of a building or occur among the
general building population but at rates similar to those found
In listing types of "sick buildings," Berglund and Lind- in buildings in general.
vall (1986) include buildings contaminated with radon, molds,
contagious agents, and formaldehyde, and they add to the 2. Building-associated illness.
Complaint or symptom rates are elevated compared to control
list "buildings in which the occupants show reactions and buildings or numbers derived from large population studies.
symptoms similar to those known to be caused by formal- Investigation confirms that the complaints are related to oc-
dehyde (Andersen et al. 1975) but in which the concentrations cupancy of the building.
of formaldehyde are well below known reaction thresholds." a) Sick building syndrome (SBS) or building sickness (BS).
Berglund and Lindvall appear to focus on defining "sick" Symptoms are similar to those identified in Tables 1 and 2,
buildings rather than the symptoms of the occupants, as in but no specific cause of the complaints can be demonstrated
SBS. Recently, Berglund and Lindvall have promoted the by the investigators.
use of the term "healthy buildings" through the convening b) Building-related illness (BAI).
of an international symposium on the subject (Berglund and The disease entity is medically identified and verified. Con-
Lindvall 1988). tamination problems determined as directly associated with
the disease(s) involve a wide range of agents and factors
Alexander and Fedoruk (1986) have categorized the par-
including infectious microorganisms, allergens, chemicals,
ticular type of problem building added by Berglund and Lind- moisture, temperature, noise, vibration and illumination.
vall as "epidemic psychogenic illness" or "mass hysteria."
c) Unclassified building associated illness.
According to Alexander and Fedoruk, the terms are inter-
changeable but the first is preferred due to the tendency to 3. Undetected problem buildings.
misunderstand the second. Like many investigations of sick
building syndrome, the diagnosis of mass psychogenic illness
is difficult to document and strongly resisted by the affected
building occupants (Alexander and Fedoruk 1986).
An example of mislabeling is a recently reported inves-
tigation of "tight building syndrome," or "closed building TABLE 5
problem," and "new building problem." The authors de- Causal factors identified in investigations of SBS (after
scribed the case as fitting the WHO (1983) SBS definition Wallace 1988).
including the usual symptoms and the failure to identify the MULTIFACTORIAL
causative agent(s) (Whorton et al. 1987). When an extended Sex
summary of the Whorton article was prepared by others later, Hay fever
the case was termed an investigation of an outbreak of "build- Migraine
Smoking
ing-related illness" (HESIS 1987).
Home-related illness
Carbonless copy paper use
Building-Related Illness Xeroxing > 25 sheets/d
Many authors- explicitly distinguish SBS from building- VDT use > 1 h/d
Unsatisfied with job
related illness (BRI), which includes allergic respiratory dis-
ease (sinusitis, tracheobronchitis, asthma, hypersensitivity PHYSICAL FACTORS
pneumonitis, and humidifier fever), skin diseases (irritant, Ventilation
Ions
allergic, and photodermatitis), irritant syndromes (carpet
Other physical causes
shampoo, formaldehyde), and infections (Legionnaire's dis-
ease, Pontiac fever, Q fever) (Kreiss and Hodgson 1984; CHEMICAL FACTORS
Formaldehyde
Hodgson and Kreiss 1986; Stolwijk 1984; Woods 1987; Other volatile organic chemicals
M0lhave 1987). Semivolatile organic chemicals
BIOLOGICAL FACTORS
Molds
Building Sickness Bacteria
I
Allergens .j
The term "building sickness" was proposed in 1984. Lars
M0lhave of Denmark suggested the term "building sickness" I
to characterize reported symptoms in a building in which "the
occupants report comfort or health problems which they as- tistically significant excess of reported SBS complaints and
sign to the indoor atmospheric environment." M0lhave would symptoms.
limit the use of the term "sick building" to cases in which the It is possible that the term "building sickness" has not
problem is identified as multifactorial and in which no mea- received more use due to the alternate meanings of the initials
sured factor exceeds generally accepted thresholds or rec- BS-British Smoke, which is used in epidemiologic studies
ommendations (M01have 1987). This is similar to most involving particulate matter, and a vulgar American slang
definitions of SBS without the criterion that there be a sta- expression.

265

.1,
~l
Etiology of Confusion building system performance or operation through either in-
M01have suggests that the different or inconsistent use spection, analysis, or environmental measurements has been
of terminology results from the involvement of different reported by these authors. No reported measurements have
groups of "indoor climate experts." For example, the term been published.
"irritation" is used by medical experts as a synonym for toxic Another large-scale British study of 30 buildings found
skin damages known from occupational exposures; by tech- "few differences in symptom prevalence" between naturally
nical and engineering experts to describe acceptability or un- and mechanically ventilated buildings. (The definition of me-
acceptability of the indoor environment; and by the occupants chanical ventilation here was "a ducted system but with no
to describe subjective feelings of reduced comfort due to dry plant for heating and cooling.") Higher symptom prevalence
nose, dry eyes, and dryness or stuffiness of the air (M0lhave rates were found for all symptoms in air-conditioned than in
1987). Thus, the backgrounds of the investigators may sig- non-air-conditioned buildings and for females than for males.
nificantly impact the diagnoses. A higher prevalence was found for all symptoms and all ven-
tilation system types in the public sector than in private-sector
Working Assumptions buildings (Hedge 1987). The published report does not detail
the inter-building variations in prevalence rates.
In the remainder of this paper, the definitions and con-
cepts articulated by the WHO Working Group (WHO 1983), The Danish investigators found no correlation between
Stolwijk (1984), M0lhave (1987), and Woods (1987) are relied complaints and building ventilation type. They found a num-
ber of factors associated with elevated symptom prevalence
upon as the basis for the discussion. These exclude building-
including age of building, sex of occupant, job category, type
related illness and emphasize multifactorial sick building syn-
drome . of work, temperature, number of occupants, amount of open
shelving, and what they called the amount of "fleecy mate-
Classification of the occupants' conditions in problem
rial" (Skov and Valbjom 1987). These findings are discussed
buildings will be classified according to a scheme adapted
in greater detail below.
from that proposed by Woods as follows: 1) unresolved build-
ing-related complaints or symptoms; and 2) building-associ- "Multifactorial Sick Building Syndrome"
ated illness including a) building sickness, b) building-related
illness, and c) unclassified building-associated illness. Woods Several investigators have suggested that the etiologic
has also added the category "undetected problem buildings" agents in SBS were multiple factors, none of which alone
which will not be discussed further here (Woods 1988). See causes increased symptoms and complaints (Turiel et al. 1983;
Table 4 for definitions of each of these categories. M0lhave 1987; Woods 1988; Valbjorn and Skov 1987). In
fact, measurement of a broad spectrum of environmental pa-
FINDINGS FROM MAJOR STUDIES rameters usually fails to isolate particular agents as etiologic.
Multifactorial analysis identifies clusters of factors associated
Potential Etiologic Agents of SBS from Recent Studies with higher rates of reported symptoms (Valbjom and Skov
Wallace has reviewed the recent literature on SBS 1987).
(1988). The review covers published reports from 1984
through 1987. The results of his review are presented in Table The Danish Town Hall Study
5. Five causal factor categories were developed: physical, The most comprehensive SBS study reported to date is a
chemical, biological, psychological, and multifactorial. Wal- multifactorial investigation known as the Danish Town Hall
lace did not review research on psychological factors. Study (DTHS). Covering a total of 27 buildings that were not
The causal factors hypothesized cover a wide range and known problem buildings , the DTHS involved measurements
the study methods also were diverse. Wallace drew no con- of indoor climate and other environmental parameters in 14
clusions from the literature, but his presentation is interesting town halls. A questionnaire and clinical study of 4369 em-
for the range of factors considered and the amount of interest ployees in the town halls and 13 affiliated buildings was also
in SBS represented by the reviewed work. conducted. While reported symptom levels were high for mu-
Several major multi-building, multi-disciplinary studies cosa! irritation (28%) and for general symptoms in the fonn
have been reported, four in Great Britain (Finnegan et al. of headache, abnormal fatigue, or malaise (36%), the mea-
1984; Pickering et al. 1985; Harrison et al. 1987; Hedge et surements of environmental parameters did not result in elu-
al. 1987) and one in Denmark (Skov and Valbjom 1987). cidation of the epidemiology.
They have developed conflicting conclusions about SBS However, the differences in the prevalence of symptoms
causal factors. among buildings was significant and was correlated with build-
Two British studies found higher symptom prevalence ing factors as well as occupant factors. Building factors that
rates in mechanically ventilated buildings than in naturally could explain the difference in the prevalence of symptoms
ventilated ones (Finnegan 1984; Harrison 1987). Among the included the total weight and potentially allergenic fraction
mechanically ventilated buildings in the later study, those of floor dust, the area of "fleecy" material per cubic meter
with humidification had the highest symptom rates. However, of air, the length of open shelves per cubic meter of air, the
among 11 naturally ventilated buildings were two in which number of work stations, and the air temperature. Occupant
symptom prevalence rates were typical of the 16 sealed build- factors included sex, type of work, and job category. See
ings in the study (Harrison et al. 1987). Table 6 for the results of the environmental measurements
Finnegan reported a threefold excess of SBS symptom (Valbjom and Skov 1987).
prevalence rates in five mechanically ventilated buildings Among the findings were the following:
(15%-45%) compared with three naturally ventilated build- 1. Elevated rates of reported mucosa) irritation were asso-
ings (5%-15%) (Finnegan and Pickering 1987). There was a ciated with the size of the allergenic fraction of floor dust,
noticeably wide range of building-specific symptom preva- the length of open shelves per cubic meter of air, the area
lence rates (5%-45%) in buildings characterized by the au- of fleecy material per cubic meter of air, the number of
thors as non-complaint buildings. Incomplete evaluation of work stations, and air temperature.

266
 - ..r. ••

TABLE 6
Indoor climate measurements in 14 Danish town halls (Valbjorn and Skov 1987).
Range
Units Mean Low High
Mean external temperature (24 hours)(°C) 2.4 -1.2 11.4
Average daily sunshine (hours) 2.3 0 6.4
Air temperature (•C) 22.7 20.5 24.1
Person-weighted air temperature (°Cl 23.0 22.0 24.4
Temperature rise during a work day ("C) 2.5 1.0 8.0
Vertical temperature gradient (°Clm) 0.9 0.4 2.0
Air velocity (mi s) 0.15 <0.15 0.20
Relative humidity (%) 32 25 40
C02 (%) 0.08 0.05 0.13
Formaldehyde mglm3 0.04 0 0.08
Static Electricity: Observer (kv) 1.4 0 4.8
Occupants max. (kv) 1.7 0 4.0
Airborne dust (mglm3) 0.201 0.086 0.382
Dust particles: > 0.5 um (1- 1) 48 x 103 19 x 1a3 119 x 1a3
> 2.0 um W1) 25 x 102 8 x 1a2 116 x 1a2
Airborne microfungi (collm 3) 32 0 111
Airborne bacteria (col/m 3) 574 120 2100
Airborne actinomycetes (collm3 ) 4 0 15
Vacuum cleaned dust" (g/12m2) 3.67 0.32 11.56
Vacuum cleaned dustl' (g/12m 2) 6.14 0.66 17.04
Macromolecular content in the dust (mg/g) 1.53 0 5.24
Macrofungi in the dust" (colf30mg) 33 11 90
Macrofungi in the dusf {col/30mg) 32 6 192
Bacteria in the dust" (colf30mg) 199 41 380
Bacteria in the dustl' (colf30mg) 296 160 680
Man-made mineral fibers in air MMMF (f/m3 ) 5 0 60
Not MMMF ( < 3 um) in the air (f X 103/m3) 0 33.2 18.5 59.1
Not MMMF (> 3 um) in the air (I x 103/m3 )d 3.1 0.7 5.0
Volatile Organic Compounds (charcoal) (mgtm3 ) 1.56 0.43 2.63
Volatile Organic Compounds (Tenax) (mg/m3) 0.5 0.1 1.2
A-weighted equivalent background (dB) 56.7 51.3 60.3
noise, [Link]
A-weighted equivalent background (dB) 36.2 28.2 44.1
noise, 4s
Reverberation time (s) 0.41 0.28 1.05

• = In the office where all lhe measurements were performed

2. Symptoms correlated strongly with job category. The vestigations that extensive measurement of environmental
symptom prevalence varied highly with job category, and variables, including air quality, may not reveal "the cause"
the highest prevalence was found in the subordinate job of the complaints. It appears more likely that a constellation
categories. Jobs involving photoprinting, working at video of factors including some combination of those identified
display terminals, and handling carbonless paper corre- above as well as chemical or biological contamination and
lated with the reported frequency of mucosa! irritation and improper ventilation system design, construction, mainte-
of general symptoms; the number of weekly working hours nance, or operation will be found in problem buildings.
of women also correlated with reports of these two symp-
tom categories, although less markedly. POTENTIAL ETIOLOGIC AGENTS IN SBS FROM
THEDTHS
3. As in several other studies, women had a higher symptom
We have hypothesized etiologic relations of some po-
prevalence rate than men and complained more frequently
tential causal agents/factors identified in the Danish Town
about indoor climate.
Hall Study (see Table 7). In Table 8 we have attempted to
4. Symptom prevalence rates varied significantly among
buildings, supporting the notion that the symptoms are identify potentially additive or synergistic co-variables. The
following discussion is based on those two tables. The dis-
building-related. Individual town halls correlated signifi-
cantly with reported mucosa! irritation and general symp- cussion focuses on some environmental and institutional fac-
toms. The lowest prevalence of symptoms was found in tors frequently associated with increased symptom or
the oldest town halls (buildings were mostly newer than complaint rates in problem buildings (Turiel 1983; M0lhave
30 years of age, with one almost 50 and another 80 years 1987; Valbjom and Skov 1987; Woods 1987).
old). VOC as Potential Sources of Complaints
5. The difference between mechanically and naturally ven- Reporting on an earlier study, M01have (1982) identified
tilated buildings was not significant for this study. This is 42 commonly used building materials and measured their vol-
in sharp contrast to the findings on the large-scale British
atile organic compound (VOC) emissions. A total of 52 com-
study (Finnegan et al. 1984; Robertson et al. 1985).
pounds were identified. An average of 22 compounds was
The DTHS strongly supports the findings of many in- identified from each material, and the range of emission rates

2fI7
 TABLE 7
Building and occupant factors (Valbjorn and Skov 1987) and their possible connection to the etiology of building
sickness.
ENVIRONMENTAL AND OCCUPANT EFFECTS
FACTORS IDENTIFIED IN HYPOTHESIZED FACTOR
DANISH TOWN HALL STUDY CO-FACTOR IN SBS ETIOLOGY
1. TEMPERATURE Microorganism proliferation Bioaerosol increase
(elevated) Higher VOC emissions/air levels voe increase
Reduced airflow Contaminant increase
Reduced ventilation Increased contaminant levels
2. FLEECY MATERIAL More voe sources Increased VOC levels
More VOC adsorption surface area Increased VOC levels
More fiber sources Increased airborne fibers
Difficult housekeeping Increased airborne dust
3. OPEN SHELVES More adsorption sites Increased VOC
Fine particle deposition sites Increased airborne particles
More difficult housekeeping Increased particles, VOC, bioaerosols
More source surface area
4. NEWER BUILDINGS More fleecy surfaces voe, particles, aerosols
< 30 years Fewer private offices Noise, crowding, contaminant levels
Higher occupant density Increased exposure to contaminants
Less occupant control Decreased resistance to illness
5. JOB CATEGORY Less mobility during day Increases exposure to contaminants
Less control over time/work Personal stress, reduced resistance
Lower status Higher exposure, less ventilation
Less control over work area Increased stress, reduced resistance
6. TYPE OF WORK More exposure to toxins More physiological stress
More exposure to irritants More physical and mental stress
Stressful work posture More physical stress
7. OCCUPANT DENSITY Lack of privacy Psychological stress
Inadequate ventilation Increased contaminant levels, exposure
More local pollutant sources Increased contaminant levels, exposure
8. #/WORK STATIONS Anonymity, impersonal environment Psychological stress
Lack of privacy, control Psychological stress
More local pollutant sources Increased exposure

was extremely large. The arithmetic average emission rate patterns of voe distribution in various locations in the two
was 9.5 mg/m2 h. Three model rooms constructed from the buildings. They concluded that the distribution of voe was
materials were found to contain between 23 and 32 of the more uniform in the healthy than in the sick preschool.
compounds at concentrations from 1.6 to 23.6 mg/m'. When
the cancer risks and health effects (M!lllhave classifies irrita- Thermal Factors
tion as a health effect) of each of the 52 compounds were Elevated temperature in a building can have many effects
reviewed, 82% were known or suspected mucous membrane on the building environment and directly or indirectly on the
irritants and 25% were suspected or known animal carcino- occupants. Not the least among them is discomfort from the
gens. temperature itself. This discomfort can reduce tolerance to
A very high percentage of common VOe emitted from other factors, many of which may be exaggerated by the
building materials are known or suspected mucous membrane elevated temperature. Additionally, microorganism growth
irritants; therefore, it is reasonable to expect significant num- may be enhanced, voe emissions from materials will in-
bers of building occupants to experience mucous membrane crease, and ventilation airflow will normally decrease.
symptoms in newly constructed, remodeled, or furnished
buildings. Turiel et al. (1981) suggested that a number of Microorganism Contamination. Biological aerosol con-
contaminants acting synergistically may have been responsi- centrations might increase due to increased growth and
ble for the higher symptom incidence in a comprehensively proliferation of microorganisms associated with higher
investigated problem building. temperatures, reduced outside airflow, and increased demand
Hollowell (1981) suggested that the reason building oc- oil air-conditioning equipment. Some organisms may prolif-
cupants complained about what they ought not to be able to erate outdoors or on building equipment surfaces in warmer
perceive (VOC) at the very low measured airborne concen- weather. An important example are the Legionella bacteria
trations was that the composite effect might give rise to the responsible for Pontiac fever, which has been reported as
reported health effects. Others have supported that theory occurring almost exclusively in spring and summer (Friedman
(M!lllhave 1982; Stolwijk 1984). et al. 1987). Many of the reported outbreaks of Legionnaire's
Noma et al. (1988) have found correlations between disease have also occurred in spring or summer.
voe profiles and their distribution patterns in one "sick" VOC Emissions. voe emissions will increase as a result
and one "healthy" Swedish preschool of identical design. of the temperature-based increase in vapor pressure. Girman
Using sophisticated statistical analyses, they examined the (1987) has calculated that a l3°e rise in temperature will result

268
 TABLE 8 presence of a greater reservoir of unattached particles which
Identification of Possible Synergistic Risk Factors for may become airborne when disturbed by human activity or
SBS Based on Risk Factors in Danish Town Hail Study ventilation airflow.
VOC Emissions. voe emitted from building materials
1. TEMPERATURE+ FLEECY MATERIAL+ OPEN SHELVES
Mo;e organic sources, emissions have been shown to distribute themselves on exposed ma-
More microbiological activity terials throughout enclosed spaces and then are re-emitted
for several weeks or more (Berglund et al. 19 7). Many
2. JOB CATEGORY+ TYPE OF WORK
Low status = loss of control, mobility, job satisfaction " fleecy" materials (such as carpets, upholstered furnishings,
Reduced proximity to windows - light, views, outside air fabric wall coverings, fiberglass insulation, and air ducts) are
Combined stressors of work posture, toxins, irritants known sources of voe. Many carpets and wall coverings are
3. NUMBER OF WORK STATIONS +AGE OF BUILDING fastened to the floors and walls , respectively , with adhesives
Higher density in open office plan, newer buildings that are known sources of voe. It is apparent that buildings
More local sources with large surface areas (from both fleecy materials and open
Less privacy, control, space shelves) will likely be associated with elevated voe air con-
4. SEX + JOB CATEGORY centrations. It is also likely that voe concentrations in such
Females in subordinate (clerical) positions buildings will decrease more slowly than in buildings with
Type of work (see #2 above) hard or smooth surfaces and less surface exposed to the in-
terior.
in a 200% increase in typical voe vapor pressure. The in-
crease in emissions will be greatest from materials with large Age of Building
surface areas in the airstream and from materials where the A clear association was found in the Danish Town Hall
emission process is dominated by evaporation from the sur- Study between age of building and complaint or symptom
face rather than diffusion from within the material. Some of rate, with the oldest buildings having the lowest rates. Newer
the materials with large exposed surface areas are freestand- offices are often constructed from softer, Jess durable mate-
ing partitions, bottoms of ceiling tiles facing the interior, tops rials on the major surfaces-floors, walls, and ceilings . This
of ceiiing panels facing concealed spaces serving as return air could result in higher airborne particle concentrations from
plenums, fibrous linings of air ducts, and textiles or fabrics deterioration of the surfaces or finishes and polishes applied
covering walls, furnishings, or floors. to them. Newer offices are usually planned with some or all
Ventilation Airflow. Many mechanical ventilation sys- "open office planning" rather than predominantly enclosed
tems will reduce airflow and outside air supply to the interior or private office spaces typical of older offices.
when temperatures rise toward the upper end of the comfort Densities of workers (per unit of area) in open offices
range. This is particularly true of variable-air-volume (VA V) are usually higher than in private offices. This results in many
systems. When temperatures are elevated, increased venti- environmental problems, including noise; chemical, physical,
lation is most necessary to remove contaminants resulting and biological contamination of air; lack of visual privacy;
from higher emission rates and increased airflow is required lack of audial privacy; and lack of control over personal work-
to provide evaporative cooling of occupants' exposed skin space.
surfaces. Yet, under most ventilation system control designs, Architects usually control open office acoustic problems
airflow or ventilation or both may be reduced. by using high-surface-area materials and components (open
shelves, freestanding partitions) and by utilizing fleecy ma-
High Surface Area Interiors terials (carpets, fibrous glass ceiling panels or insulation, fab-
"Fleecy material" and "open shelves" were identified as ric-covered partitions). This reduces reverberation time and
risk factors in the Danish Town Hall Study. "Fleecy" refers breaks direct paths of sound transmission. The partitions,
to materials such as fabrics and carpets which have rough, where used, also provide some visual privacy and a feeling
textured surfaces. The association with elevated symptoms of occupant control (at least over the immediate work station
might result from the extremely large effective surface areas area). However, they do substantially increase surface area
facing the interior space on fleecy materials or open shelving and impede space air distribution.
systems. "Effective" surface area refers to the actual surface
area available for adsorption and re-emission sites for voe Job Category and Type of Work
or deposition of small particles. Recent advances in mathe- Lower status jobs were associated with higher complaint
matics (fractal math) have shown that the actual surface area and symptom rates in the DTHS. Subordinate workers, such
available for particle deposition or molecular adsorption is as clerical and drafting personnel, tend to spend more time
many times larger than the two-dimensional (plane geometry) at the work stations than their supervisors, who often move
measured surface area. Small particles, which penetrate deep- about the building or leave it to attend meetings. Lower status
est in the respiratory system, deposit equally efficiently on workers also tend to have Jess space and to be located near
horizontal and on vertical surfaces (Weschler 1988). the interior of the building.
Carpets, textiles used for wall coverings and furnishings, Outside air supply to the interior is usually less than at
and insulation materials facing the interior or the airstream the perimeter. Some buildings deliver primarily or only re-
in mechanical ventilation systems are high-surface-area ma- circulated air to the interior spaces. Thus, interior spaces may
terials that provide more adsorption sites for voe and more have stagnant or stale air. The combination of higher density,
deposition sites for small particles. Fibrous materials also pro- more activity, and less ventilation would resuit in higher con-
vide readily available source material for airborne particles centrations of airborne contaminants. Perimeter offices would
through surface erosion, abrasion, or deterioration of the be more likely to have views out of windows, providing the
binding forces. Housekeeping tasks including cleaning, vac- worker with visual and psychological relief.
uuming. and dusting are made more difficult by rough sur- Type of Work. Certain jobs involve exposure to chem-
faces and the larger surface areas, thereby resulting in the ical and physical agents known to cause irritation, nervous

269
system effects, and other health outcomes. Equipment and considered necessary and it is not economically feasible to
materials used in duplicating, printing, mailing, and clerical identify causes if remedial measures can effectively reduce
activities in general are all associated with various chemical complaints and symptoms, as in the case of ventilation system
and physical agents which may contaminate indoor air. It is modifications . Thorough characterization of environmental
also likely that individuals performing such work will be in factors, including detailed chemical and biological contami-
subordinate positions and therefore at risk as discussed nant measurements, can be prohibitively expensive and is
above. only undertaken in the most severe or persistent of cases.
Occupant Density and Number of Work Stations Protocols to guide investigations of problem buildings
have not been widely tested, validated , or promulgated by
It is not clear whether the DTHS found high occupant
any standards development organization. While some gen-
density associated with higher symptom report rates because
eralized protocol have been prepared, significant differences
of the density or the possible correlation of high density with
among problem building cases require individualization of
job category and type of work, as described above. Lower
protocols and measurement methods (Levin 1987c; NAS
status workers are likely to have less assigned personal work
1985; NIOSH 1987; Sterling et al. 1987; Woods et al. 1988).
area, i.e., higher occupant density. At higher densities, the
A standard guide for investigation of problem buildings is
occupant-generated air contaminants (metabolic- and activ-
currently being prepared by the ASTM (Levin 1988).
ity-based) will be more concentrated prior to dilution or re-
Standardized sampling and analytical methods for indoor
moval by ventilation.
air are limited and those that exist are not uniformly applied
A higher number of work stations might be associated
with increased anonymity, a lack of personal privacy and or widely used. Efforts to address these shortcomings are
control, and a higher rate of contaminant generation. All of under way by the ASTM (Levin 1988). However, interbuild-
these could affect occupant stress and comfort levels. ing variations limit standardization of investigations and com-
Woods (1988) has found a high proportion of buildings parability of results (Levin 1987c).
with high occupant densities that exceed the design capacities Comprehensive monitoring for airborne contaminants is
of the ventilation system. Occupants are added without mod- extremely expensive and rarely definitive in problem building
ification to the ventilation system, resulting in inadequate air investigations. Monitoring for airborne VOC is expensive and
there is a lack of general agreement regarding appropriate
supply, interruptions of design airflows, and excessive loads
monitoring methods. Characterization of total VOC is not as
on cooling and heating equipment.
expensive as identification and quantification of specific com-
Lll\flTATIONS OF INVESTIGATIONS pounds, although the methodological problems are signifi-
There are numerous limitations on investigations of cant.
problem buildings which suggest explanations for the fre- Even where extensive environmental monitoring is con-
quent failure to identify etiologic agents or contributory ducted, interpretation of results is limited by the absence of
building factors . Among these limitations are cost, timeliness guidelines and standards. Interpretation is often based upon
investigatory methods, building complexity, building dynam- standards or guidelines developed in and for different con-
ics, institutional constraints, and insufficient guidance to in- texts, such as the industrial workplace and ambient air. Com-
vestigators. parison of measurements to such guidelines or standards can
Unsystematic and incomplete investigations result in in- lead to incorrect assumptions about the effect of the measured
adequate diagnoses and unsuccessful remedial efforts. Yet parameter on occupant health and comfort (Eisinger 1988).
complete, systematic investigations are rare. Expert investi-
gators are usually selected by the occupants or building owner Limitations Imposed by the Problem Context
according to the owner's perceptions of the problem etiology. Buildings are dynamic, responding to changes in the ex-
Expertise is usually confined to one or a limited number of ternal environment, internal loads, and the building itself.
fields, resulting in incomplete investigations and narrowly fo- Internal loads may be generated by user activities, building
cused findings in many (M0lhave 1987; Kreiss and Hodgson equipment, or occupants' appliances and equipment. Build-
1984). ing loads vary as a result of normal operation of building
The perseverance of the investigators and the availability equipment, principally lighting, ventilation, heating, and
of methods can determine whether chemical and biological cooling. Malfunctions in building equipment, interventions
agents can be eliminated as etiologie of building-associated by occupants, manipulation by building operators, and signals
outbreaks (Kreiss and Hodgson 1984). In many instances, from building systems controls constantly effect changes in
ventilation system problems are identified early in the inves- equipment operation, resulting in load changes and load han-
tigation. Modifications to the system equipment, operating dling.
schedule , or operational modes (airflow, temperature) will Environmental variations among locations within a single
result in a significant reduction of complaints and symptoms, building can be enormous. Small distances between locations
and the investigation will be terminated before problem can involve large differences in some environmental variables
causes are defined. including critical air quality factors. Significant variations in
Timely investigations rarely occur due to institutional environmental factors occur hourly, daily, and seasonally·
constraints. Frequently, when complaints or symptoms are Therefore, monitoring of environmental factors, including
initially reported, there is hesitation by management to give but not limited to the sampling and analysis of indoor air,
importance to them. It is often only when complaints become can produce misleading results unless an adequate number
very numerous, when upper-level management personnel are of representative samples is collected over extended or ad-
affected, or when workers initiate organized or formal action equately representative time periods and locations. This may
that management commissions investigators. be an especially important factor in the failure of many large
Comprehensive and systematic investigations are expen- building or multi-building studies where associations between
sive, difficult, and more time-consuming than building op- reported symptoms and environmental variables are not
erators or users can normally tolerate. In many cases it is not found at statistically significant levels. Single-area measure-

270
 ments or one measurement per floor or per ventilation system Most advocates of phased diagnostic investigations of
for many occupants will simply average precisely the varia- problem buildings urge extremely limited use of airborne
tions that the study seeks to elucidate. monitoring during the initial phase. They assert that the ma-
Designer, operator, and occupant perceptions frequently jority of building problems can be solved or resolved without
differ from each other and from actual building conditions. extensive monitoring. Furthermore, it is argu·:!d that moni-
Investigator interviews with some but not all of these parties toring is of limited effectiveness until it can be focused on
can result in biased assumptions or hypotheses followed by hypothesized causal agents or factors.
incomplete or poorly focused investigations. 2. Qualitative diagnostics-hypotheses are formulated
through engineering analysis; system performance analysis is
DISCUSSION OF SICK BUILDING DIAGNOSES initiated with limited measurements (such as airflow and pres-
AND FINDINGS sure differences) . Medical evaluation identifies suspect pol-
By definition , an SBS diagnosis requires confirmation lutants and air or bulk samples will be collected for these
that an elevated complaint or symptom rate is associated with substances.
occupancy of a particular building, that no known etiology 3. Quantitative diagnostics-if further investigation is
accounts for the symptoms, and that clinical evidence of needed to test hypotheses, samples will be collected and ana-
building-related illness is absent. lyzed and other environmental measurements will be made.
SBS can be hypothesized but not diagnosed or defined
based solely on clinical evidence unless sufficient numbers of
BUILDING ECOLOGY
occupants are examined in the case building and compared
to occupants of a control building or to a valid baseline for l ndoor air quality is begin ning to receive recognition as
symptom prevalence. Clinical identification of symptoms as- an importanc indoor environmental factor as lighting, acous-
sociated [Link] occupancy of a particular [Link] can result in tics, privacy, security, thermal comfort, and aesthetics have
an SBS hypothesis. Only an epidemiologic or clinical inves- received historically . An app roach to understanding buildings
tigation can define the occurrence of SBS. Even where epi- and human health based on a systematic, comprehensive
demiologic evidence supports an SBS hypothesis, BRT may framework is badly needed . We recommended an ecological
be present in some or all of the occupants manifesting SBS approach-a methodology utilizing knowledge and analytical
symptoms. methods like those used by biological scientists in the study
SBS may involve diverse reactions among building oc- of living organisms in relationship to their environment. We
cupants. This may be a function of varying conditions within have borrowed from the co re of the definition of ecology to
the building, varying individual responses to environmental coin the term "building ecology, " which we define as the
factors, or both. Where this is the case, it wiJI reduce statistical tudy of buildings and their relationship to the natural and
associations between reported symptoms and measured en- built environment around them and to humans who use or
vironmental conditions, and may lead to incorrect interpre- are otherwise affected by them (Levin 1981 ).
tations of even the most complete and careful investigations. We suggest that the concepts of dynamic, interdependent
Therefore , as defined and discussed in this paper and flows used in studying ecosystems exemplify methods that can
elsewhere, SBS may not be a useful term in that it refers to be adapted to the study of indoor air and human health . An
symptom sets which are manifestations of various distinct example is a mass balance and mass flow analysis of a con-
illnesses or diseases. Aggregating these distinct medical con- taminant or of moisture into through, and out of a building.
ditions may be the greatest barrier to discovery of the causes Models for such mass balances and flows have shown that
of sick building syndrome. changes in one factor can shift the rate of the processes and
the overall distribution of contaminants. Chaos theory in
Building Diagnostics phy ics has shown that a small perturbation can initiate major
r "Building diagnostics" is the name given to a set of prac-
tices used to assess the current performance and capability
deviations from a steady state or regular periodic behavior.
Building environmental control systems designed to address
of a building and to predict its likely performance in the future these perturbations are themselves subject to changes re-
(NAS 1985). Whi le building diagnostics can be valuable at sulting in small perturbations. There are time constants or
many stages in the life of a building, it may be most useful lags for each of these changes as well. Thus , the system is an
in investigations of problem [Link]. Four elements are es- ever-shifting collection of interconnected entities-in the case
[Link] to building diagnostics, according to the NAS report; of the building, both living and inanimate; in the case of the
they are as follows: ecosystems, each organism and its environment.
(1) knowledge of what to measure,
(2) availability of appropriate instruments and other PREVENTING AND REMEDIATING PROBLEM
measurement tools, BUILDINGS
(3) expertise in interpreting the measurements, and
Based on the causal facrors identified in the Danish Town
(4) capability of predicting the future condition of the
Hall Study and other investigations, we have identified some
building based on that interpretation.
potential preventive or remedial measures to minimize SBS.
Several authors have proposed phased investigations of
These measures, which are listed in Table 9, have not been
problem buildings or in ocher applications of building diag-
systematically evaluated, but there is considerable evidence
nostics {NAS 1985; Sterling et al. 1987; NIOSH 1987· Woods
to support their potential efficacy in reducing the occurrence
et al. 1988). Woods et al. (1988) have divided the phases as
of sick building syndrome.
described below.
l. Consultation-scope of the investigation is defined
and observations of the building and its systems are made CONCLUSION
(walkthrough ·urvey). Few or no instrumented measurements Buildings are complex. Their effects on humans are ex-
are made. tensive and are poorly understood. Attitudes toward and un-

271
 TABLE 9 fems; IAQ '88. Atlanta: American Society of Heating,
Refrigerating, and Air-Conditioning Engineers. Inc.
Potential Control Measures for Multifactorial SBS
Berglund, B., and Lindvall , T. 1986. "Sensory reaction to
HARD SURFACES 'sick' buildings. " Env. Int., Vol. 12, pp. L47-l59 .
Reduce surface area of materials exposed to interior, Berglund, B., and Lindvall , T. L988 (Eds.). Healthy buildings
Use building form, layout, and surface treatment for acoustic
control. '88, vols. 1-4. Stockholm: Swedish Council for Building
Research.
LOWEST COMFORTABLE AIR TEMPERATURE Berglund B.; Johansson, I. · and Lindvall, T. 1987. "Volatile
E$peclally when building or furnishings are new or after floor and organic compounds from building materials in a simu-
furniture polishing, waxing. lated chamber study." Indoor Air '87: Proceedings of the
MINIMIZE OPEN STORAGE SHELVING 4th lmernational Conference 011 Indoor Air Quality and
Use enclosed shelves where possible. Climate, Vol. 1. Berlin: lnstitute for Water , Soil and Air
Locate shelves in separate space with exhaust ventilation, no re- Hygiene, pp. 16-21.
circulation. Colligan, M.J. 1981. "The psychological effects of indoor air
LOW OCCUPANT DENSITY pollution." Bull. N. Y. Acad. Med., Vol. 57, pp.
Utllize total building space for roughly equal area per occupant. 1014-1025.
Avoid crowding. Confirm HVAC capacity for additional occupants. Eisinger, D.; Mudarri , D.; and Austin, B. 1988. "New policy
challenges, limited analyrical tools: using existing air
MAXIMIZE OUTDOOR AIR SUPPLY quality health standards to diagnose indoor air quality
Extend hours and days of operation; increase percent outside air.
Start-up earlier after days of vacancy. problems." Presented to the Association for Public Pol-
icy Analysis and Management Tenth Annual Research
INCREASE WORKER CONTROL, PRIVACY Conference, October 27 , Seanle, WA.
Provide local and individual control over lighting, ventilation, heat- Fanger, P.O. 1987. " A solution to the sick building mystery."
ing, cooling, acoustic environment. Indoor Air '87: Proceedings of the 4th International Con-
FLEXIBLE WORK HOURS ference on Indoor Air Quality and Climate. Vol. 4. West
Allow individual schedule, where feasible.
Berlin , 17-21 August. Berlin: Institute for Water, Soil
and Air Hygiene.
MINIMIZE EXPOSURE IN STRESSFUL JOBS Finnegan, M.J.; Pickering, C.A.C.; and Burge, P.S. 1984.
Rotate jobs, mandatory rest (fresh air?) breaks. "The sick building syndrome: prevalance studies .. , Brit.
Med. J., Vol. 289, pp. 1573-1575.
Finnegan , M.J., and Pickering, C.A.C. 1987 . · Prevalence of
symptoms of the sick building syndrome in buildings
derstandings of buildings' effects on occupant health and without expressed dissatisfaction ." Indoor Air '87: Pro-
comfort are not generally shared. ceedings of th e 4th international Conference on indoor
Sick building syndrome is inadequately understood at Air Quality and Climate. Vol. 4. West Berlin, 17-21 Au-
this time, in part due to confusion in definitions and termi- gust. Berlin: Institute for Water, Soil and Air Hygiene
nology, in part due to inadequate efforts to study compre- pp. 542-546.
hensively its occurrence and causes. Potential causes have Friedman , S.; Spitalny, K. ; Barbaree, J.; Faur , Y.; and
been identified and discussed, but the definition of SBS may McKinney , R. 1987. " Pontiac fever outbreak associated
itself preclude elucidation of its causes due to the Limitations with a cooling tower." American Joim1al of Public
on investigations. There are great difficulties inherent in con- Health, May Vol. 77 No . 5, pp. 568-572.
ducting comprehensive investigations or srudies that are suf- Girman, J. 1987. "Bake-out of an office building. " Indoor
ficiently sensitive to detect relevant associations. Air '87: Proceedings of the 4th lntemational Conference
Causes of SBS require further investigation. An ap- on Indoor Air Quality and Climate, Vol. l. West Berlin ,
proach for conceptualizing and conducting problem building 17-21 August. Berlin: Institute for Water , Soil and Air
investigations has been outlined. We recommend that further Hygiene, pp. 22-26.
efforts to develop and refine models for application in di- Harri on , J.; Pickering A.C.; Finnegan M.J.; and Austwick,
P. K. C. 1987. •'The sick building syndrome: further prev-
agnosing problem buildings be developed. Field studies, lab-
alence studies and investigation of possible causes." In-
oratory studies, and modeling efforts need to be performed
door Air '87: Proceedings of the 4th International
to further elucidate the causes and nature of sick building Conference 011 Indoor Air Quality and Climate, Vol. 2.
syndrome. West Berlin, 17-21 August. Berlin: Institute for Water,
Potential control measures have been identified. Eval- Soil and Air Hygiene . pp. 487-491.
uation of the efficacy of potential control measures is war- Hedge, A. ; Wilson, S. ; Burge, P.S.; Robertson , A .S.; and
ranted due to the large costs involved in their implementation Harris-Bass , J. 1987. ' Indoor climate and employee
and the risks attendant to their failure. health in offices." Indoor Air '87: Proceedings of the 4th
lmemational Conference on Indoor Air Quality and Cli-
REFERENCES mate, Vol. 2. West Berlin, 17-21 August. Berlin: Insti-
Alexander , R.W ., and Fedoruk, J. 1976. "Epidemic psycho- tute for Water, Soil and Air Hygiene, pp. 492-496.
genic illness in a telephone operator's building. 'J. Occ. HESIS. 1987. "Investigation and work up of tight building
Med., Vol. 28 No. 1, pp. 42-45. syndrome. The state of the work place," Vol. l, No : 2.
Alsbirk , K.E.; Johansen, M.; and Petersen. R. 1983. " Ocular An internal bulletin from HESIS, the Hazard Evaluauon
symptoms and exposure co mineral fibres in boards for System and Information Service, California Department
ound insulation of ceilings." Ugeskr. Laeger. Vol. 145 of Health Services, Berkeley, p. 12. .•
pp. 43-47 (English summary). [Link] on, M.J . 1988 . "Health risks of indoor pollutants.
Andersen , l. ; Lundqvist, G.R. ; and M01have, L. 1975. "In- Engineering Solutions 10 Indoor Air Problems: IAQ '88.
door air pollution due to chipboard used a a construction April 11-13. Atlanta: American Society of Heating. Re-
material." Atmospheric Environment, Vol. 9, pp. frigerating, and Air-Conditioning Engineers, Inc., PP·
1121-1127. 284-293.
A~HRAE. 1988. Engineering solutions to indoor air prob- Hodgson, M.J. 1989. Personal communication to author.

272
 ·1
('
Hodgson , M.J. , and Kreiss, K. 1986. " Building associated Morey , P.R.; Jones, W.G. · Clere J.L. ; and Sorenson, W.G. I,
diseases." Proceedings of the ASHRAE Conference l 986a. "Studies on sources of airborne microorganisms 'I
IAQ'86: Managing Indoor Air for Health and Energy and on indoor air quality in a large office building." ii
Conservation . April 20- 23. Atlanta: American Society Indoor Air '86: Managing Indoor Air for Health and
of Heating , Refrigerating, and Air-Conditioning Engi- Energy Conservation. Atlanta: American Society of
neers, lnc. , pp. 1-15. Heating , Refrigerating and Air-Conditioning Engi-
neers Inc. , pp. 500-509.
Hollowell, C.D., and Miksch, R.R. 1981. "Sources and con-
centrations of organic compounds in indoor environ- Morey, P.R. et al. 1986b. "Environmental studies in moldy
ments." Bull. N. Y. Acad. Med., Vol. 57, pp. 962-977. office buildings." ASHRAE Transactions, Vol. 92, Part
l, pp. 399-419.
Kreiss, K., and Hodgson, M. 1984. "Building-associated ep- Nagda, N.; Rector, H.; and Koontz, M. 1986. Guidelines for
idemics." In: Walsh, P.J.; Dudney, C.S.; and Copen- monitoring indoor air quality. New York: Hemisphere
haver, E.D. (Eds.), Indoor Air Quality, pp. 87-106. Publishing Corp.
Boca Raton, FL: CRC Press, Inc. NAS. 1985. Building diagnostics: a conceptual framework.
Levin, H. 1981. "Building ecology." Progressive Architecture, Washington , DC: National Academy Press.
Vol. 62, No. 4, pp. 173-175. NIOSH. 1987. Guidance for indoor air quality investigations.
Levin , H. 1987a. "What architects can do to improve indoor Cincinnati: National Institute for Occupational Safety
air quality. " Indoor Air '87: Proceedings of the 4th In- and Health.
ternational Conference 011 Indoor Air Quality and Cli- Noma, E.; Berglund, B.; Berglund, U.; Johansson, I.; and
mate, Vol. 4, Plenary Lectures. West Berlin, 17- 21 Baird, J. 1988. "Joint representation of physical loca-
August. Berlin: Institute for Water, Soil and Air Hy- tions and volatile organic compounds in indoor air from
giene , pp. 17- 26. a healthy and a sick building." Atmospheric Environ-
ment, Vol. 22, No . 3, pp. 451-460.
Levin, H. 1987b. ' A procedure used to evaluate building Norback, D.; Rand, G.; and Michel, I. 1987. "Overfrequen-
materials and . furnishings for a large office building." cies of skin and airway symptoms among personnel in
Indoor Air '87: Proceedings of the 4th lmernational Con- sick buildings and polluted industrial environments in
ference on Indoor Air Quality and Climate, Vol. 1. West relation to exposure." Indoor Air '87: Proceedings of the
Berlin, 17-21 August. Berlin: Institute for Water, Soil 4th International Conference on Indoor Air Quality and
and Air Hygiene, pp. 54-58. Climate, Vol. 2. West Berlin 17- 21 August. Berlin: Jn-
Levin, H. 1987c. " Overview of indoor air quality sampling stitute for Water , Soil and Air Hygiene , pp. 515- 519.
and analysis. " ASTM Special Technical Publication 957, Robertson, A.S.; Burge, P.S .; Hedge, A.; Sims, J.; Gill,
Sampling and Calibration of Atmospheric Measurements. F.S.; Finnegan, M.; Pickering, C.A.C.; and Dalton, G.
Philadelphia: American Society for Testing and Mate- 1985. "Comparison of health problems related to work
rials. and environmental measurements in two office buildings
Levin, H. 1988. "ASTM Subcommittee D22.05 on Indoor with different ventilation systems." Br. Med. J., Vol.
Air; Progress of work to develop standard methods for 291, pp. 373-376.
sampling and analysis of indoor air" (Paper 88-94.5). Samet, J.M.; Marbury, M.C.; and Spengler, J.C. 1988.
Presented at APCA Annual Meeting, Dallas, TX, June ' Health effects and sources of indoor air pollution , part
20-24. II." Am. Rev. Respir. Dis., Vol. 137, pp. 221-242.
Levin, H., and Phillips, T.J. 1989. "Pre-occupancy meaure- Skov, P. ; Valbjom O.; and DISG. 1987. "The 'sick' building
ments in California state office buildings." Design and syndrome in the office environment: the Danish town
Protocol for Monitoring Indoor Air Quality, ASTM STP hall study." Env. Int., Vol. 13, pp. 339-349.
1002. Nagda, N.L., and Harper, J.P., Eds. Philadelphia: Stolwijk, J.A.J. 1984. "The 'sick building' syndrome." In:
American Society for Testing and Materials, pp. 99-110. Berglund, B.; Berglund, V.; Lindvall, T.; and Sundell,
J. (Eds.), Indoor Air: Proceedings of the 3rd International
Lyon, M.H. 19--. "Microbial problems in air conditioning." Conference on Indoor Air Quality and Climate, Vol. 1,
The Heating and Ventilating Engineer, Vol. 58, No. 674, Recent Advances in the Health Sciences and Technol-
pp. 8-11. ogy. Stockholm: Swedish Council for Building Research,
M0lhave, L. 1982. "Indoor air pollution due to organic gases pp. 23-29.
and vapours of solvents in building materials." Env. Int., Turiel, I.; Hollowell, C.D.; Miksch, R.R.; Rudy, J.V.;
Vol. 8, pp. 117-127. Young, R.A.; and Coye, M.J. 1983. "The effects of re-
duced ventilation on indoor air qulaity in an office build-
M0lhave, L. 1984. Unpublished workshop at the 3rd Inter- ing." Atmospheric Environment, Vol. 17, No. l, pp.
national Conference on Indoor Air Quality and Climate,
51-64.
Stockholm.
M0lhave, L. 1987. "The sick buildings-a subpopulation Wallace, L. 1988. "The sick building syndrome: a review."
among the problem buildings?" Indoor Air '87: Pro- 81st Annual Meeting of APCA, Dallas, TX, June 19-24.
ceedings of the 4th International Conference on Indoor Weschler, C.J., and Shields, N.C. 1988. "The influence of
Air Quality and Climate, Vol. 2. West Berlin, 17-21 Au- HVAC operations on the concentration of indoor air-
gust. Berlin: Institute for Water, Soil and Air Hygiene, borne particles." IAQ 88: Engineering Solutions to In-
pp. 469-473. door Air Problems, pp. 166-181.
Morey, P.R. 1988. "Microorganisms in buildings and HV AC Whorton, M.D.; Larson, S.R.; Gordon, N.J.; and Morgan,
systems: a summary of 21 environmental studies." En- R. W. 1987. "Investigation and work-up of tight building
gineering Solutions to Indoor Air Problems: IAQ '88. syndrome." J. Occup. M., Vol. 29, No. 2, pp. 142-147.
Atlanta, April 11-13. Atlanta: American Society of Woods, J.E. 1987. Testimony presented before the Subcom-
Heating, Refrigerating, and Air-Conditioning Engi- mittee on Environmental Protection of the Committee
neers, Inc., pp. 10-24. on Environment and Public Works, U.S. Senate, No-
Morey, P.R. Undated. "Sources of microorganisms and cor- vember 20.
rective actions that may be useful in reducing microbial Woods, J.E. 1988. "Recent developments for heating, cool-
contamination in buildings." Golden Valley, MN: Ho- ing and ventilating buildings; trends for assuring healthy
neywell Indoor Air Quality Diagnostics. buildings." In: Berglund, B., and Lindvall, T. (eds.),

273
 . ,- , - . . ., . .
. .· . :·~
. ... -:~
. ,......".

Heahhy Buildings '88: Vol. 1, State of the Art Reviews. ing of the Working Group on Assessment and
Stockholm: Swedish Council for Building Research, Monitoring of Exposure to Indoor Pollutants. Copen-
June, pp. 99-107. hagen: World Health Organization, Regional Office for
Europe.
Woods, J.E.; Morey, P.R.; and Stolwijk, J.A.J. 1988. "In- WHO. 1986. "Indoor air quality research." EURO Reports
door air quality and the sick building syndrome: a di- and Studies 103. Report on a meeting of the Working
agnostic approach." Applied Industrial Hygiene, in press. Group on Indoor Air Quality Research, Stockholm. Co- ·
WHO. 1983. "Indoor air pollutants: exposure and health ef- penhagen: World Health Organization, Regional Office
fects." EURO Reports and Studies 78. Report on a meet- for Europe.

}
I
•

}
;
I

- -'Zf.-4 -----------~