There is a well-known discrepancy between dosimetrically derived dose conversion factor (DCF) and epidemiologically derived DCF for radon. As the latter DCFs, International Commission on Radiological Protection (ICRP) recommends a value of ∼6.4 nSv (Bq h m(-3))(-1) and 7.9 nSv (Bq h m(-3))(-1) for radon decay products (RnDP) in dwellings and workplaces, respectively. On the other hand, the dosimetric calculations based on the ICRP-66 respiratory tract model derived a DCF of 13 nSv (Bq h m(-3))(-1) and 17 nSv (Bq h m(-3))(-1) for RnDP in dwellings and workplaces, respectively, and 83 nSv (Bq h m(-3))(-1) for thoron decay products (TnDP) in dwellings. In addition, the DCFs derived from both approaches and UNSCEAR were applied to comparative dosimetry for two thoron-enhanced areas (cave dwellings in China and dwellings at a spa town in Japan), where the equilibrium equivalent concentration of radon and equilibrium equivalent concentration of thoron have been measured. In the case of the spa town dwellings, the dose from TnDP was larger than the dose from RnDP.