Here are two abstracts on the subject from Medline. The first one is the original article on which the New Scientist article is based. The second is an earlier study by German researchers with some interesting results.
Effect of ultraviolet germicidal lights installed in office ventilation systems on workers' health and wellbeing: double-blind multiple crossover trial
Dick Menzies, Julia Popa, James A Hanley, Thomas Rand, Donald K Milton
--------------------------------------------------------------------------------
Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute (D Menzies MD, J Popa MEng) and Department of Epidemiology and Biostatistics, McGill University, Montreal, Canada (D Menzies MD, J A Hanley PhD); Department of Biology, St Mary's University, Halifax, Nova Scotia, Canada (T Rand PhD); and Department of Environmental Health, Harvard School of Public Health, Boston, MA, USA (D Milton PhD)
--------------------------------------------------------------------------------
Correspondence to: Dr Dick Menzies, Respiratory Epidemiology Unit, Montreal Chest Institute, McGill University, 1110 Pine Avenue West, Room 103, Montreal, Quebec, Canada H3A 1A3 (e-mail:dick.menzies@mcgill.ca)
Summary
Background Workers in modern office buildings frequently have unexplained work-related symptoms or combinations of symptoms. We assessed whether ultraviolet germicidal irradiation (UVGI) of drip pans and cooling coils within ventilation systems of office buildings would reduce microbial contamination, and thus occupants' work-related symptoms.
Methods We undertook a double blind, multiple crossover trial of 771 participants. In office buildings in Montreal, Canada, UVGI was alternately off for 12 weeks, then turned on for 4 weeks. We did this three times with UVGI on and three times with it off, for 48 consecutive weeks. Primary outcomes of self-reported work-related symptoms, and secondary outcomes of endotoxin and viable microbial concentrations in air and on surfaces, and other environmental covariates were measured six times.
Findings Operation of UVGI resulted in 99% (95% CI 67-100) reduction of microbial and endotoxin concentrations on irradiated surfaces within the ventilation systems. 771 participants appeared to remain masked, and reported no adverse effects. On the basis of within-person estimates, use of UVGI was associated with significantly fewer work-related symptoms overall (adjusted odds ratio 0·8 [95% CI 0·7-0·99]), as well as respiratory (0·6 [0·4-0·9]) and mucosal (0·7 [0·6-0·9]) symptoms than was non-use. Reduction of work-related mucosal symptoms was greatest among atopic workers (0·6 [0·5-0·8]), and never-smokers (0·7 [0·5-0·9]). With UVGI on, never-smokers also had large reduction of work-related respiratory (0·4 [0·2-0·9]), and musculoskeletal symptoms (0·5 [0·3-0·9]).
Interpretation: Installation of UGVI in most North American offices could resolve work-related symptoms in about 4 million employees, caused by microbial contamination of heating, ventilation, and air-conditioning systems. The cost of UVGI installation could in the long run prove cost-effective compared with the yearly losses from absence because of building-related illness.
Lancet 2003; 362: 1785-91
Zentralbl Hyg Umweltmed. 1997 Aug;200(2-3):137-51.
[Sick building syndrome and HVAC system: MVOC from air filters]
[Article in German]
Schleibinger HW, Wurm D, Moritz M, Bock R, Ruden H.
Institut fur Hygiene der FU Berlin.
Growth and emissions of volatile metabolites of microorganisms on air filters are suspected to contribute to health complaints in ventilated rooms. To prove the microbiological production of volatile organic compounds (MVOC), concentrations of aldehydes and ketones were determined in two large HVAC systems. The in situ derivated aldehydes and ketones (as 2,4-dinitrophenyl-hydrazones) were analysed by HPLC and UV detection. The detection limit of each compound was 1 ppb (margin of error < 10%). Field measurements were carried out before and after the prefilters and the main filters, respectively, to investigate whether aldehydes and ketones increase in concentration after filters of HVAC systems. First results show that the compounds formaldehyde, acetaldehyde and acetone could be detected before and after the filters. The concentrations of these VOC after the filters were significantly increased--as a mean over twenty measurements--, especially as far as filters made of glass fibre are concerned. However the found concentrations were low and mostly comparable to outdoor findings. In simultaneous laboratory experiments pieces of used filter material of one HVAC system and unused filter pieces (for blank values) were examined in small incubation chambers to investigate the possible production of MVOC. For the incubation a temperature of 20 degrees C and a relative humidity of 95% was chosen. In these experiments an almost identical spectrum of compounds (formaldehyde and acetone) was found as in the field measurements. The concentrations of these compounds were higher in the chambers with the used filter pieces. The concentration of acetone ranged up to almost 12 mg/m3.--As our field experiments correspond with our laboratory experiments, we assume that the microbial production of volatile organic compounds in HVAC systems under operating conditions is possible.