Shelter and indoor air.

Improvements in outdoor air quality that were achieved through the implementation of the Clean Air Act accentuate the quality of the indoor air as an important, if not dominant, factor in the determination of the total population exposure to air contaminants. A number of developments are adding important new determinants of indoor air quality. Energy conservation strategies require reductions in infiltration of outdoor air into buildings. New materials introduced in the construction and in the maintenance of buildings are contributing new air contaminants into the building atmosphere. Larger buildings require more and more complex ventilation systems that are less and less under the individual control of the occupants. All of these factors contribute to the current reality that indoor air contains more pollutants, and often at higher concentrations, than outdoor air. Especially in the larger buildings, it will be necessary to assure that an adequate quantity of fresh air of acceptable quality is provided to each individual space, and that no new sources of pollutants are added to a space or a whole building without appropriate adjustments in the supply of fresh air.


Introduction
There are complex interactions between indoor air quality, indoor climate, and other conditions of occupancy in residential buildings and nonindustrial workplaces that result in nonspecific complaints and concerns. Such complaints can sometimes be associated with the growth of microorganisms in a building and its systems. Health, the quality of life, and ultimately productivity of substantial segments of the population are affected in ways and to extents that are currently poorly described and quantified. Since a very large proportion of our daily lives is spent in various forms of shelter, an even larger proportion of our total exposure to a large number of air pollutants is determined by the building environment.
We actually spend, on average, 85 to 90% of our 24-hr day in some form of shelter, be it a home, a car, an office, school, or workplace. The shelter provides us with a microenvironment with an optimized temperature and protection from sun, wind, and precipitation. In the days of heavy outdoor air pollution our shelters also provided us with some protection from the peaks ofthat pollution. In the last few decades the outdoor air and the industrial workplace have attained much lower levels of air pollution as a result ofthe activities under the Clean Air Act and the Occupational Safety and Health Act.
In recent years the nonoccupational indoor environment is receiving an increased level of attention, and in some form or another, this is likely to continue and increase well into the twenty-first century. Future developments in this area are likely to be shaped by trends *Department of Epidemiology and Public Health, and John B. Pierce Foundation Laboratory, Yale University School of Medicine, New Haven, CT 06510. that are already discernable: * Energy will be progressively more costly, and less flexible. * Shelters will be larger buildings with higher densities of occupants because of energy considerations, limitations in available land, and in transportation systems. * The materials surrounding us in our shelters will be more and more ofmanufactured synthetic rather than natural origin. * The fraction of the population spending a large part of their 24-hr day in large, mechanically ventilated buildings will continue to increase; the total annual population exposure to complex building environments will therefore also increase. * The current tendency towards diminishing individual control of personal environments in large buildings will continue. Twenty years ago the quality of indoor air in the residential and the nonindustrial occupational environment was not seen as an important issue, and it was assumed to be adequate. Since then, a number of studies in the United States, West Germany, Italy, and the Netherlands have collected data on pollutant levels in several thousand residences (1)(2)(3)(4). A working group of the World Health Organization reviewed these data (5) and concluded that in all these industrialized nations, the same large number of contaminants occurred in the residential environment in about the same concentrations and in the same distribution of concentrations. As might be expected, the concentrations are quite variable over space and time. For the majority of the pollutants examined, the concentrations indoors were higher or much higher than the outdoor concentrations, indicating that they were due to sources within the shelter. These insights have important consequences for pubair. In addition, the background incidence rate and the lic health and for strategies for reduction in total popu-annual U.S. mortality attributed to leukemia is given. lation exposures to a whole range ofair pollutants. Table The aggregate of the TEAM observations on benzene is 1 presents the annual air intake via the respiratory route given in the first six columns of Table 2. The last two and the annual and lifetime intake of benzene and tolcolumns provide the calculated attributable incidence of uene from the indoor and outdoor environment. The leukemia resulting from the exposures described in the benzene and toluene concentrations in Table 1 are taken first six columns, based on the population and unit risk from the median values reported in (5).
numbers in the header. The form in which the existing It is clear that the dominant intake is from the indoor exposure data and risk projections are given in Table 2 environment, and that also makes the indoor environ-allows for the evaluation of the effectiveness of different ment the most effective target for attempts to reduce exposure reduction strategies. the total population exposure. Another way in which we A similar projection can be made for the distribution of can evaluate the distribution of such indoor exposures is the risks of lung cancer attributable to indoor radon presented in Table 2. This table presents the output daughter concentrations as described in the U.S. (6). from a spreadsheet that can be used to assess the conse- Table 3 presents such a spreadsheet for radon daughquences that can be expected from the distribution of ters. Exposures to benzene and radon would not genbenzene exposure over the population, given the carci-erally lead to acute effects, nor would these exposures nogenic potency estimates for benzene. lead to recognition of an odor, except perhaps at the Table 2 presents a comprehensive assessment of the highest concentration in Table 2. health consequences of the distribution of benzene ex-There have been occasions in which formaldehyde was posures, which was established in the Total Exposure introduced into residential environments from inap-Assessment Methodology (TEAM) study (1), relating it propriately formulated or installed urea formaldehyde to the carcinogenic potency estimate developed by the foam insulation or from inappropriately fabricated chip-Carcinogen Assessment Group (CAG) of the U.S. Enviboard. The rate of complaints involving formaldehyde in ronmental Protection Agency (EPA) and to the threshresidential environments is now at a very much lower old limit value (TLV) for benzene. In addition, Table 2 level than was experienced at the time of initial introplaces these outcomes into the total perspective of the duction of these products. total leukemia incidence in the United States. The foot-Occupants in a large number of buildings in the U.S. note of this table contains the U.S. population, the TLV and in other industrialized nations have complained for benzene, and the unit risk per year per million people about acute adverse effects associated with their presof leukemia per microgram per m of benzene in inhaled ence in the buildings. These conditions have been re-  ferred to as the sick building syndrome, or the tight or stuffy building syndrome. They have been reported in dayeare centers of kindergarten schools in Sweden (7), in large apartment buildings in Denmark, and in office buildings in the United Kingdom and the United States. In all cases the occupant complaints have a great deal of similarity that has I to the characterization as a syndrome. The symptoms that are characteristically reported in excessive frequency are as follows: irritation of eyes, nose, and throat; headaches and dizziness; odors; fatigue and lethargy; wheezing and sinus congestion; skin rash and irritation; and nausea. All these symptoms are reported with a 10 to 20% background incidence in any population, and it is not a simple matter to determine what the minimum incidence in a population should be and whether or not any reported incidence among the occupants in a given building at any time is significantly different from that minimum achievable incidence.
It is clear that an excess incidence of the symptoms previously discussed will have an effect on the productivity of an office population, but at the present time there are not any quantitative measures of such an effect. The effects are usually acute and reversible after leaving the offending building environment, and the complaints are usually limited to a minority of occupants.
When buildings that have given rise to occupant complaints are investigated along the lines ofan occupational hazard evaluation, it is unusual to find a particular pollutant that is present in sufficient concentration to account for the occupant complaints. In a large number of such investigations the National Institute for Occupational Safety and Health (NIOSH) investigators found that inadequate ventilation was the most common cause identified (8). The most logical conclusion would then be that the inadequacy ofthe ventilation causes a number of contaminants to rise in concentration at the same time, and that it is the total concentration of the contaminants that is responsible for the occupant complaints, rather than the presence of any single contaminant in a critical concentration. Molhave (9) in Denmark has carried out experimental exposures ofhuman volunteers to complex mixtures of organic air pollutants in which he reported responses at concentrations which for each of the constituents would be below the threshold for perception. Little is known about the sensitization of some individuals at such low concentrations to render them more sensitive than the remainder of the population.
Ventilation systems and their components are often capable of supporting substantial growth of microorganisms such as fungi, algae, and bacteria. Such growth can occur in cooling towers, in the ventilation system itself in spray humidification systems, and in cooling coils. Microorganisms can be distributed via the ventilation air stream from the system to the occupied spaces, where sensitive occupants can be severely affected in reactions ranging from irritation to Legionellosis and Pontiac fever.
The systems supporting large buildings have become quite complex. A typical large office building might have 43 floors, each with 20,000 ft2 of floor space. The ventilation air for the lower 20 floors is supplied by a machine floor near the 21st floor, and the remaining floors are served by another floor of blowers near the top floor. That means that air is distributed from a central location to each of 20,000 ft2 on each of 20 floors. The air travels a long way through stacks and ducts both to and from a particular office. Even if the distribution was once perfect, it soon is disturbed by incompetent adjustment or by changes in the use of space or in the occupant density. Attempts to correct a local problem often create problems in other locations because of the interaction between these adjustments. If ductwork or equipment is inappropriately designed, installed, or maintained, it can support growth of microorganisms that get distributed throughout the building. Occupants can become sensitized to such microorganisms or to products from such growth. Some of the organisms are directly pathogenic such as Legionella. It is rare that all the equipment and ducts in a large building are kept clean with any consistency, and often they are not even accessible for maintenance and cleaning.
The foregoing discussion indicates that the ventilation air supply in a large building with a tight envelope is of critical importance. The quality ofthe air supplied can be degraded by microbiological growth in the system, by contamination ofthe intake air from vehicle exhausts, or from the cooling tower drift. When the rate ofsupply to a space is inadequate, the contaminants released in the space are not diluted enough. Occupants have no direct control over either the quantity or the quality of ventilation air; as a result, they are likely to seek redress ifthe indoor air quality is felt to be less than adequate. In the twenty-first century we are likely to have to control the air quality in buildings as closely as the quality of the drinking water that is being supplied to the occupants. The supply of ventilation air in large buildings is likely to take on the characteristics of a regulated activity such as the potable water supply or the supply of safe and healthful food. It is also likely that there will be increased scrutiny of building materials, furnishings, maintenance, and cleaning products introduced into building spaces. It is not possible to design and operate ventilation systems in buildings that can deal effectively with the sudden and very large sources of contaminants that are regularly introduced whenever pesticides must be used or when wall or floor coverings are glued to the structure.
All of the foregoing discussion applies to residential environments with equal relevance. There are major differences between nonindustrial workplaces and residences: in a residence the density of occupants is usually much lower than in office buildings, and occupants in residential environments have a substantial amount of individual control over their immediate environment. In a residence the occupants determine whether and when pollutants are intoduced, and they can open windows.
The current revision of the American Society of Heating, Refrigerating and Air-Conditioning Engineers Ventilation Standard 82-1989(10) is attempting to deal as effectively as possible with current and future needs of architects, engineers, contractors, and owners and operators of buildings, but the state of our knowledge is quite inadequate to do much more than specifying empirically derived rates at which outdoor air must be supplied to interior spaces. Knowledge about the effects of indoor air pollutants on building occupants is quite inadequate to specify safe and acceptable levels in indoor spaces.
Regulations are perhaps not an effective approach to achieve improvements in indoor air quality, but interest by labor unions and the rapidly growing interest in the legal profession in tort actions on behalf of building occupants are likely to focus increasing attention on the problem of indoor air quality in public access buildings. At the present time we cannot estimate the economic leverage of indoor air quality in office buildings, but it does not require a complex calculation to show that even a very small effect on the productivity of office workers would justify a substantial research effort in the area of indoor air quality and also a substantial increase in cost of operation and maintenance of ventilation systems.