Sick buildings and the experimental approach

Abstract When dealing with indoor air pollution problems caused by substances known to be toxic, field experimentation has been virtually ruled out by ethical considerations. Even the omission of remedial measures for experimental purposes can be said to be unethical. What remains is the descriptive approach, in which the occurrence of toxic substances and the occurrence of symptoms of ill health are measured and related to each other. This approach is ethically unexceptional, but leaves much to be desired in terms of scientific proof, unless the mechanisms of toxicity are well understood. When applied to the elusive Sick Building Syndrome (SBS), it is fatally flawed: the toxic substances involved, and the mechanisms for their action, are not known, and the sub‐clinical symptoms are usually measurable only as subjective complaints. In descriptive surveys, complaints can be manipulated by alarmist reports and by deliberate campaigns; they may even be used by subjects merely as a vote for something to be done. This is not the case in properly conducted field experiments, where subjects know that they may well be part of the reference group. Subjective assessments of symptom intensity may therefore be used as valid dependent variables, and the extent to which they are affected by experimentally established conditions may be examined. Provided that the experimental conditions may or may not be beneficial, the experimental approach is ethically defensible. The recent series of field experiments at Malmö General Hospital (MAS) in Sweden is used to demonstrate the power of the approach.