Minireview

How Much Protection Against Radon Do We Need?

Klaus Becker

CEJOEM 2001, Vol.7. Nos.3-4.:168-177

It has been statistically demonstrated with miners in various countries, including Germany and Bohemia, during a long period reaching from the Middle Ages to the early or middle 1950s, that the radioactive decay products of radon constitute a strong contributing factor to an increased lung cancer rate. A “mountain disease” affecting the lungs of silver miners in the Ore Mountains (Erzgebirge) of Southern Saxony close to the Czech border was first mentioned in 1537 by the famous physician Paracelsus (the same who first stated that “it is the dose which makes the poison”). This miner’s disease, later also known as “Schneeberg lung cancer” a fter a mining town in this area, was identified as lung cancer 120 years ago, and originally attributed to the high arsenic concentrations in the mine air (for review, see Becker, 2000).
      In 1913, the increase in lung cancer incidence was first related to the extremely high radon levels, exceeding in some cases several MBq/m³ (corresponding to several Sv/y according to ICRP Recommendations 65 and 66, containing values that differ by a factor of four within a single year between the two issues, or four orders of magnitude above the intervention level of 148 Bq/m³ currently recommended for homes by the U.S. Environmental Protection Agency). Radon was certainly an important contributing factor to the increase of lung cancer rate in a still poorly understood synergism, apparently ranking the 4th after smoking, silica, and arsenic inhalation. Other contributing factors were uranium and other mineral dusts, nitrous gases, diesel exhaust fumes, etc., to which the miners on the German and Czech sides of the border had been exposed during the Klondike-like conditions of the uranium mining boom in the first years after World War II.
      On the German side, 250–300 thousand miners extracted 220,000 t of uranium between 1945 and 1989 for the Soviet nuclear program. However, since the early 1950s conditions in the mines substantially improved due to better ventilation, wet d rilling, etc. Since Germany’s reunification, ca. 7 billions Euro of public funds have been spent in this region for remediation measures, approximately one third of which was related to overground radon evaluation and reduction programs (Becker, 1996).
      Obviously, conditions in early mining cannot be extrapolated down over many orders of magnitude to the residential radon situation. Indeed, no detrimental health effects have been observed among the rather stable population not only regarding lung cancer, but also concerning other solid cancers and leukaemia (Conrady et al., 2001). Some of the old houses in the Schneeberg area have been inhabited by the same families for many generations. Up to 115,000 Bq/m³ were measured, and 12% of all homes exceeded 15,000 Bq/m³ (or 100 times the US limits), thus providing a “natural laboratory” for radon studies. In particular, it was interesting to note that lung cancer, although it was easy to diagnose in that miners’ environment, remained extremely rare in the non-miner population before mass consumption of cigarettes started.

One example may demonstrate the predominant effect of smoking. Among the approx. 20,000 autopsies performed since 1852 in one of Saxony`s largest hospitals in the capital Dresden, the percentage of lung cancers among all reasons of death in the autopsies slowly increased during the proceeding decades. It rose from only 0.06% between 1852 and 1876 to 0.21% from 1877 to 1884. Between 1885 and 1894, it was 0.43% (Schüttmann, 1999). And it is important to note that in 1862, Germany`s first cigarette factory had started its production in Dresden. WHO estimates that by now almost half of exogenous cancers are caused by smoking world-wide.
      It has been well established that in “retrospective smoking dosimetry” smokers notoriously underestimate their past and present smoking habits (even more than drug and alcohol addicts), in particular after lung cancer has been diagnosed (Conrady et al., 2001). Thus, this factor by far prevails over other problems in residential radon epidemiology, such as large uncertainties in long-term retrospective radon dosimetry, even if they may deeply influence the results. In particular, the early radon dosimetry in uranium mines was rather unreliable or absent. Even today, radon concentrations in a room have been shown to fluctuate by a factor up to 500 depending on ventilation and other factors. Structural changes over years are known to substantially influence radon levels, and changing construction codes are likely to have a serious impact on the further development of residential radon levels.
      There have been many attempts to establish an equivalent between the assumed risks of residential radon and smoking habits. During a Biophysics Colloquium in the radon spa Bad Schlema near Schneeberg in 1993, top European radon researchers quoted numbers between 500 WLM equivalent to 1 pack of cigarettes/day and as little as one cigarette per day (thus reaching the realm of passive smoking) corresponding to an assumed doubling risk of 400 Bq/m³, which is almost twice that of the currently discussed 250 Bq/m³ limit for new buildings in the European Union. Obviously, a misjudgement of only one cigarette per day may thus falsify most case-control studies, unless they are restricted to never-smokers. Typically, in such studies the number of the ever-smokers among the lung cancer cases is much larger than that among the controls, e.g., in the Iowa Lung Cancer Study 86% of the lung cancer cases were smokers, but only 32% of the controls (Lubin and Boice, 1997).

For a non-epidemiologist, the multitude of published results are confusing. The most frequently quoted meta-analysis of case-control studies compiles the very different results of nine such studies in seven countries (Lubin and Boice, 1997). Of about 30 data points with very large vertical error bars, but no horizontal bars relating to the rather difficult retrospective radon dosimetry, only one from Sweden is slightly above the control line (Fig. 1). Even this single indication of a slight lung cancer increase at 450 Bq/m³ has been recently shown to be transferable into a decrease (Lagarde and Pershagen, 1997) similarly to those of the well-known and widely discussed large-scale ecological studies in the USA when the same type of evaluation was used (Cohen, 1995).
      In the U.K. and Canada there are other meta-analysis studies in progress, which will incorporate the results of more recent case-control studies, performed, e.g., in the U.K. (Darby et al., 1998) and USA (Field et al., 2000). In the high radon areas of former East Germany, one government-funded 10 million-dollar study reports a slightly positive trend, while another, sponsored by the EU and restricted to non-smoking females, finds clear indications of an initially negative trend and a threshold around 1000 Bq/m³ (Conrady et al., 2001; Fig. 2). A most obvious factor for the explanation of such confusing data appears to be cigarette smoking.

      Such investigations raise a number of interesting questions. Even the BEIR VI report does not exclude the possibility of a threshold value (Mossman, 1998). There are resonable mechanistic models consistent with an U-shaped dose-response relationship for radon (Bogen, 1998), and it had been concluded by Mossman (1998) that “epidemiological evidence to support (EPA) conclusions is either absent or not convincing. A more reasonable conclusion is that lung cancer risk is insignificant for radon concentrations below 400 Bq/m³. Radon does not pose a threat to the public health in the domestic environment.” This confirms a suspicion which was already expressed over a decade ago, that “with radon an artificial disease has been created by the multiplication of a very small risk large populations, in order to obtain frightening numbers” (Letourneau, 1987).

There are many research data which directly or indirectly support conclusions, of which only a few can be listed here as examples:

(1)	Because of various factors which have so far been largely neglected, such as various aerosol size and daughter product attachment factors, as well as additional radiation factors such as the inhalation of Rn-220 (thoron) decay products, long-lived radioactive dust, and external gamma radiation, the actual lung exposures of underground miners are likely to be much higher (up to a factor of 5–10) than previously estimated, which implies an overestimate of residential radon effects by the same risk factor.
(2)	The bronchial alpha dose attributed to radon progeny has been subject to many modifications in the past, and still substantially varies depending on the lung model used.
(3)	The relative risk of lung cancer for external low-LET exposures decreases, contrary to the LNT hypothesis, before reaching a threshold around 2 Gy (Rossi and Zaider, 1997; Fig. 3).
(4)	Experimental animal studies demonstrate a RBE for lung cancer induction by alpha radiation of 2 instead of 20 as assumed by ICRP Recommendation 60, possibly replaced in the next edition by 10 (Kellington et al., 1997).
(5)	There is experimental evidence for a definitely non-linear radon effect, amounting to an overestimation of residential radon risks, which has recently been provided in single alpha particle exposures of mammalian cells (Miller, 1999).
(6)	Regarding incorporated alpha emitters such as Ra und Pu in humans and animals, there is substantial evidence that no effect is detectable up to at least 2–4 Gy, as demonstrated, for example, in careful long-time studies of about 1700 female radium dial painters, with a threshold for osteosarcoma induction at 10 Gy corresponding to 200 Sv according to ICRP 60 (Rowland, 1995; Fig. 4).
(7)	Animal experiments demonstrate that it is more the radon inhalation dose-rate than the integrated dose which determines the lung cancer induction (Monchaud and Morlier, 2001). Moreover, radiobiological evidence accumulates demonstrating basic differences in radiobiological response between low and high doses and dose-rates.

      Bearing particular relevance, the substantial historical evidence of beneficial effects of radon on human health should not be forgotten, even if there could be some overlap of beneficial effects dominating under some, and slightly detrimental effects under other circumstances, a phenomenon well-known in pharmacology. In fact, radon may be one of mankind`s oldest therapies. The ancient Romans and other old civilizations appreciated radon spas; in Japan, the springs on Misasa (up to 160,000 Bq/l of radon in drinking water) have been used for 800 years.
      Currently, in each year about 75,000 patient are treated in German and Austrian radon spas, and many more in other countries, in particular in Russia, mostly for painful joint or backbone diseases such as rheumatic arthritis and ankylosing spondylitis (morbus Bechterew). The treatment is by inhaling high radon concentrations (in the Bad Gastein “Heilstollen”, for example, about 170,000 Bq/m<sup>3</sup>), by drinking, or by bathing in radon water. The expenses (e. g., in Bad Gastein about $500 for a 10-hour treatment) are mostly paid by public health insurance systems. Even in the more radiophobic USA, a “radon health mine” has been operating successfully for half a century.
      Clinical studies, including randomized double-blind ones, clearly demonstrated the superior effect of radon in comparison with an otherwise identical treatment, lasting for several months after the end of the treatment. Various mechanisms have been suggested as explanation of such beneficial effects involving radiation doses in the order of only one mGy, including stimulation of repair or radical scavenger enzymes, or the production of neuropeptides, but there are still many open questions. However, considering the practical success and acceptance of radon treatments by the patients suffering from severe pains, radon balneology has recently been accepted (similar to other treatments for which there is not yet a profound scientific explanation) by German authorities. Thus it is not surprising that the radon medical treatment center of Bad Schlema, which had been very popular up to the end of World War II as the “world`s strongest radon source”, was officially reopened a few years ago and is again, like the nearby old “St. Joachimstal”, very popular with health-seeking patients.

The current residential radon “dangers” may be summarized as follows (Becker, 2000; Becker and Schüttmann, 1998):

(1)	The regulatory policy is inconsistent, with suggested intervention levels fluctuating between countries by fairly large factors, and ICRP recommending a radon limit for the population which is by a factor of 10 larger than for external gamma radiation.
(2)	The underlying linear extrapolation from high mining exposures to low levels in homes appears to be not feasible because of significantly differing confounding factors and dose levels, and different radiobiological response mechanisms at high and low levels.
(3)	Epidemiological studies indicating small increase of relative risk for increased residential levels appear to be subject to large errors, in particular in the retrospective determination of smoking habits.
(4)	There is evidence for a threshold or a U-shaped response curve, perhaps explaining some of the beneficial effects of radon balneology. Up to about 1000 Bq/m3, lung cancer appears to be attributable to smoking only.
(5)	In all but very few exceptional cases, such as extremely high radon levels in the homes of heavy smokers, the cost/benefit ratio of residential radon programs does justify the investment of private or public funds.

BECKER, K. (1996). “Wie teuer ist uns Radon – und weshalb?” Atomwirtschaft, Atomtechnik 41:108–109.

BECKER, K. (2000). “Is residential radon dangerous?” In: The Effects of Low and Very Low Doses of Ionizing Radiation on Human Health. Excerpta Med. Int. Congr. Ser. 1203. Elsevier, Amsterdam, pp. 173–191.

BECKER, K. and FIELD, R.W. (2001). “Debate: How dangerous is residential radon?” Radiat. Protect. Dosim. in press.

BECKER, K. and SCHÜTTMANN, W. (1998). “Was ist eigentlich aus dem Radon geworden?” Strahlenschutzpraxis 4:54–58.

BOGEN, K. T. (1998). “Mechanistic model predicts a U-shaped relation of radon exposure and lung cancer risk.” Human Exp. Toxicol. 17:691–696.

COHEN, B. L. (1995). “Test of the linear no-threshold theory of radiation cancerogenesis for inhaled radon decay products.” Health Phys. 68:157–174.

CONRADY, J., MARTIN, K., LEMBCKE, J., and MARTIN, H. (2001). “The true size of lung cancer risk from indoor radon: Hidden behind a smoke screen?” Excerpta Med. Int. Congr. Series. Elsevier, in press.

DARBY, S., WHITLEY, E., SILCOCKS, P., THRAKRAR, B., GREEN, M., LOMAS, P., MILES, J., REEVES, G., FEARN, T., and DOLL, R. (1998). “Risk of lung cancer associated with residential radon in SW England: A case-control study.” Brit. J. Cancer 78:394–408.

FIELD, R. W., STECK, D. J., SMITH, B. J., BRUCE, C. P., NEUBERGER, J. S., FISHER, E. L., PLATZ, C. E., ROBINSON, R. A., WOOLSON, E. R., and LYNCH, C. F. (2000). “Residential radon gas exposure – The Iowa Lung Cancer Study.” Am. J. Epidemiol. 151:1091–1102.

KELLINGTON, J. P., ELDRED, T. M., AMBROSE, K., BROOKS, P. M., and PRIEST, N. D. (1997). “Effects of radiation quality on the lung cancer in CBA/Ca mice.” In: Proc. Int. Conf. Health Effects of Low Level Radiation, Stratford-on-Avon. BNED, London, pp. 44–51.

LAGARDE, F. and PERSHAGEN, L. (1997). “Residential radon and lung cancer in Sweden: Risk analysis accounting for random error in the exposure assessment.” Health Phys. 72:269–276.

LETOURNEAU, E. G. (1987). Ref. in How serious is the indoor radon health hazard? JAMA 258:578.

LUBIN, J. H. and BOICE, J. D. (1997). “Lung cancer risk from residential radon: Metaanalysis eight epidemiological studies.” J. Natl. Cancer Inst. 89:49–57.

MILLER, R. C. (1999). “The oncogenic transformation potential of the passage of single alpha particles through mammalian cell nuclei.” Proc. Natl. Acad. Sci. USA 96:19–22.

MONCHAUD, G. and MORLIER, J. P. (2001). “Influence of dose-rate on lung cancer induction in rats exposed to radon and its progeny.” In: Excerpta Med. Int. Congr. Ser. 1225. Elsevier, Amsterdam, in press.

MOSSMANN, K. L. (1998). “Is indoor radon a public health hazard? The BEIR VI Report.” Radiat. Prot. Dos. 80:357–360.

ROSSI, H. H. and ZAIDER, M (1997). “Radiogenic lung cancer: The effects of low doses of low linear energy transfer (LET).” Radiat. Environ. Biophys. 36:85–88.

ROWLAND, R. E. (1995). “Dose-response relationship for female Ra dial painters: a new look.” In: Health Effects of Internally Deposited Radionuclides. World Scientific, London, pp. 135–143.

SCHÜTTMANN, W. (1999). “Bewertung des Lungenkrebsrisikos durch Wohnungsradon.” Strahlenschutzpraxis 5:35–40.