A test of the linear-no threshold theory of radiation carcinogenesis

It has been pointed out that, while an ecological study cannot determine whether radon causes lung cancer, it can test the validity of a linear-no threshold relationship between them. The linear-no threshold theory predicts a substantial positive correlation between the average radon exposure in various counties and their lung cancer mortality rates. Data on living areas of houses in 411 counties from all parts of the United States exhibit, rather, a substantial negative correlation with the slopes of the lines of regression differing from zero by 10 and 7 standard deviations for males and females, respectively, and from the positive slope predicted by the theory by at least 16 and 12 standard deviations. When the data are segmented into 23 groups of states or into 7 regions of the country, the predominantly negative slopes and correlations persist, applying to 18 of the 23 state groups and 6 of the 7 regions. Five state-sponsored studies are analyzed, and four of these give a strong negative slope (the other gives a weak positive slope, in agreement with our data for that state). A strong negative slope is also obtained in our data on basements in 253 counties. A random selection-no charge study of 39 high and low lung cancer counties (+4 low population states) gives a much stronger negative correlation. When nine potential confounding factors are included in a multiple linear regression analysis, the discrepancy with theory is reduced only to 12 and 8.5 standard deviations for males and females, respectively. When the data are segmented into four groups by population, the multiple regression vs radon level gives a strong negative slope for each of the four groups. Other considerations are introduced to reduce the discrepancy, but it remains very substantial. Since cigarette sales data are available only on a statewide basis, mean radon data for states are analyzed. The linear regression for lung cancer rates vs radon levels is also negative and has a much steeper slope than that for the county data. When cigarette sales per capita is introduced into the regression, the negative slope for dependence on radon level is essentially unchanged.     

Epidemiological associations among lung cancer,radon exposure and elevation above sea level--a reassessment of Cohen''s county level radon study

Inhalation of radon (222Rn) decay products by persons living in homes has been associated with increased risk of lung cancer. Some epidemiological studies have shown a positive association between radon exposure and lung cancer rates. However, a large U.S.-wide ecological study (Cohen 1995) has shown a clear inverse association between average county radon concentration in homes and average lung cancer rates in the county. Cohen's strong inverse association between radon and lung cancer is surprising since there is no plausible biological reason for an inverse causal relationship between the two. We plot the county average lung cancer rate vs. the elevation above sea level (altitude) and show an inverse association between county average lung cancer rate and elevation. The elevation used for each county is the altitude of the most populous place in the county. We postulate that the decrease in lung cancer rates with higher elevations is caused by the carcinogenic effect of higher absolute oxygen concentration in the inspired air at lower elevations. Stratifying Cohen's lung cancer vs. radon data into ten groups of counties with similar elevations removes some, but not all, of his inverse association between radon and lung cancer.     

Residential radon exposure and lung cancer in Swedish women.

A case-control study was undertaken to investigate the role of residential radon exposure for lung cancer. The study included 210 women with lung cancer diagnosed from 1983-1986 in the county of Stockholm and 191 hospital and 209 population controls. Interviews provided information on lifetime residences and smoking. Radon concentrations measured in 1,573 residences of the study subjects showed a lognormal distribution with arithmetic and geometric means of 127.7 and 96.0 Bq m-3, respectively. Lung cancer risks tended to increase with estimated radon exposure, reaching a relative risk of 1.7 (95% confidence interval: 1.0-2.9) in women having an average radon level exceeding 150 Bq m-3 (4 pCi L-1). Stronger associations were suggested in younger persons and risk estimates appeared to be within the same range as those projected for miners. However, further studies are needed to clarify the level of risk associated with exposure to residential radon.     

Smoking as a confounder in ecologic correlations of cancer mortality rates with average county radon levels

Cohen has reported a negative correlation between lung cancer mortality and average radon levels by county. In this paper, the correlation of U.S. county mortality rates for various types of cancers during the period 1970-1994 with Cohen's radon measurements is examined. In general, quantitatively similar, strongly negative correlations are found for cancers strongly linked to cigarette smoking, weaker negative correlations are found for cancers moderately increased by smoking, whereas no such correlation is found for cancers not linked to smoking. The results indicate that the negative trend previously reported for lung cancer can be largely accounted for by a negative correlation between smoking and radon levels across counties. Hence, the observed ecological correlation provides no substantial evidence for a protective effect of low level radon exposure.     

Selenium in forage crops and cancer mortality in U.S. counties

The potential protective effect of selenium status on the risk of developing cancer has been examined in animal and epidemiologic studies. This ecological study investigated the association between U.S. county forage selenium status and site- and sex-specific county cancer mortality rates (1950-1969) using weighted least squares regression. Consistent, significant (p less than .01) inverse associations were observed for cancers of the lung, rectum, bladder, esophagus, and cervix in a model limited to rural counties and for cancers of the lung, breast, rectum, bladder, esophagus, and corpus uteri in a model of all counties. No consistent significant positive associations were observed in the rural county models. This remarkable degree of consistency for the inverse associations strengthens the likelihood of a causal relationship between low selenium status and an increased risk of cancer mortality.     

Respiratory organs cancer mortality in New Jersey counties and the relationship with selected demographic and environmental variables

The state of New Jersey (N.J.) has been thought to have an unusually high overall cancer mortality rate; this assumption has been based on national 1950–1969 mortality data for N.J. counties. This study presents an analysis of more recent rates of respiratory cancer mortality in 21 N.J. counties during 1968–1977, a comparison with the 1950–1969 rates, and associations between current respiratory cancer mortality rates and selected demographic and environmental variables. Age-adjusted mortality rates for cancer of respiratory organs were calculated for the N.J. counties during the period 1968–1977 and compared with the period 1950–1969, with the Surveillance, Epidemiology, and End Results (SEER) survey, and with cancer mortality in the United States, 1973–1977. The county rates were also correlated with chemical toxic-waste disposal sites (CTWDS), annual per capita income, percentage of the population employed in chemical industries, the density of population, and the urbanization index of each of 21 N.J. counties. The lung, bronchus, trachea, and pleura cancer mortality rates among white and nonwhite males and females in N.J. were substantially higher than the national rates during the period 1950–1969. In more recent years, the increases in U.S. mortality rates for lung, bronchus, trachea, and pleura cancers were significantly greater (P < 0.01) than those found in most of the 21 N.J. counties. As a consequence, the national rates are now more comparable to N.J. rates. Although the gaps between N.J. and the United States in these rates have narrowed, the observed number of laryngeal and lung cancer deaths remained significantly higher (P < 0.01 to P < 0.0001) than expected cancer deaths, based on U.S. rates, among one or more subgroup populations (white and nonwhite males and females) in several N.J. counties. Among white men in Middlesex, Camden, Burlington, and Ocean counties, the observed number of deaths for lung cancer was found to be significantly (P < 0.0001) greater than the expected number of deaths. In Hudson county observed deaths from both laryngeal and lung cancer among white men were significantly greater than the expected number of deaths from these cancers (P < 0.0001). Statistically significant and positive correlations were found between laryngeal cancer mortality and CTWDS, urbanization index, and population density. Lung cancer mortality also correlated significantly with CTWDS in N.J. Both larynx and lung cancer mortality showed significant and consistent negative correlations with annual per capita-income in N.J. Some of the implications of the study findings are discussed and recommendations made for future investigations.     

Association of lung cancer mortality with precambrian granite

Sixteen counties in New York, Pennsylvania, and New Jersey that are associated with the Reading Prong granite deposits have significantly higher age-adjusted lung cancer rates among whites of both sexes (1950-1979) than do 17 nearby control counties. Elevated radon daughter concentrations have been found in homes near the Reading Prong granites. Fraction of populations living in cities with over 5,000 persons, industrial centers, cities with populations above 20,000, and median incomes did not differ significantly for three county groups (those which include the granite, fringe area, and control areas). Weaknesses were inadequate home measurements of radon and lack of smoking information. Findings are consistent with several other studies relating radon in homes to lung cancer.     

History of residence and lung cancer risk in New Mexico

Participants in a population-based case-control study of lung cancer in New Mexico between 1980 and 1982 were asked to identify all locations where they had resided for six months or more. These residential data were coded at the county and state levels and combined with county-level socioeconomic data from the 1910, 1930, 1950, and 1970 decennial censuses to generate indices of time lived in counties or metropolitan areas of different sizes, degrees of urbanization, or extents of employment in manufacturing industries. Urban residence was not associated with employment of male controls in jobs or industries considered to increase lung cancer risk. However, in the non-Hispanic white female controls, urban residence before age 30 years in a county of 500,000 or more residents was associated with a fourfold higher odds ratio for starting to smoke cigarettes. Male and female non-Hispanic controls who had ever lived in more populous counties smoked more cigarettes per day than did those who had not lived in such counties. Residential history patterns were the same in cases and controls; multiple logistic regression showed no consistent associations of the residence history variables with lung cancer risk.     

Bladder cancer mortality in New Jersey counties, and relationship with selected environmental variables

Age-adjusted mortality rates for bladder cancer were calculated for the 21 New Jersey (NJ) counties (USA) during the period 1968-1977, and compared with the period 1950-1969, with the Surveillance, Epidemiology and End Results (SEER) survey and with cancer mortality in the US 1973-1977. The county rates were also correlated with: the rates of low birth weight, birth defects, infant mortality; chemical waste disposal sites; annual per capital income; per cent of the population working in the chemical industries; density of population and urbanization indices of 21 NJ counties. Age-adjusted bladder cancer mortality rates in 95% of NJ counties were higher than national and SEER area rates. The overall NJ State rates for four subgroup populations were highly significantly (p less than 0.001) greater than the national rates. There was a statistically significant correlation between bladder and lung cancer mortality among females in 21 NJ counties which may suggest a common risk factor--namely cigarette smoking. There was no such correlation between bladder and lung cancer mortality among males. There was a statistically significant association between bladder cancer mortality in individual counties and the percentage of the adult population working in the chemical industries.     

The potential for bias in Cohen's ecological analysis of lung cancer and residential radon.

Cohen's ecological analysis of US lung cancer mortality rates and mean county radon concentration shows decreasing mortality rates with increasing radon concentration (Cohen 1995 Health Phys. 68 157-74). The results prompted his rejection of the linear-no-threshold (LNT) model for radon and lung cancer. Although several authors have demonstrated that risk patterns in ecological analyses provide no inferential value for assessment of risk to individuals, Cohen advances two arguments in a recent response to Darby and Doll (2000 J. Radiol. Prot. 20 221-2) who suggest Cohen's results are and will always be burdened by the ecological fallacy. Cohen asserts that the ecological fallacy does not apply when testing the LNT model, for which average exposure determines average risk, and that the influence of confounding factors is obviated by the use of large numbers of stratification variables. These assertions are erroneous. Average dose determines average risk only for models which are linear in all covariates, in which case ecological analyses are valid. However, lung cancer risk and radon exposure, while linear in the relative risk, are not linearly related to the scale of absolute risk, and thus Cohen's rejection of the LNT model is based on a false premise of linearity. In addition, it is demonstrated that the deleterious association for radon and lung cancer observed in residential and miner studies is consistent with negative trends from ecological studies, of the type described by Cohen.     

Expected indoor 222Rn levels in counties with very high and very low lung cancer rates

Counties in the U.S. with high lung cancer rates should have higher average 222Rn levels than counties with low lung cancer rates, assuming the average 222Rn level in a county is not correlated with other factors that cause lung cancer. The magnitude of this effect was calculated, using the absolute risk model, the relative risk model, and an intermediate model, for females who died in 1950-1969. The results were similar for all three models. We concluded that, ignoring migration, the average Rn level in the highest lung cancer counties should be about three times higher than in the lowest lung cancer counties according to the theory. Preliminary data are presented indicating that the situation is quite the opposite: The average Rn level in the highest lung cancer counties was only about one-half that in the lowest lung cancer counties.     

Environmental carcinogen releases and lung cancer mortality in rural-urban areas of the United States

Purpose: Environmental hazards are unevenly distributed across communities and populations; however, little is known about the distribution of environmental carcinogenic pollutants and lung cancer risk across populations defined by race, sex, and rural‐urban setting. Methods: We used the Toxics Release Inventory (TRI) database to conduct an ecological study at the county level (a total of 3,141 counties). Multiple linear regressions were used to assess the associations of carcinogenic discharges from TRI sites and lung cancer mortality rates at the county level in the United States during the years 1990 through 2007. Findings: We observed an excess risk of population lung cancer mortality associated with higher amounts of environmental carcinogen releases from TRI facilities in both males and females, and in both whites and African Americans. The strength of these associations tended to be stronger in African Americans. A significant dose‐response relationship was observed for the total volume of carcinogen releases or carcinogen releases to the air, but not releases to water. These associations appeared to be present within nonmetropolitan counties but not metropolitan counties, and to be concentrated in certain urban‐rural county typologies. Conclusions: Our results suggest that exposure to higher carcinogen releases from industrial or chemical facilities in rural areas may increase the risk of lung cancer mortality. Our findings add to the evidence for undertaking prudent efforts to limit the release of carcinogenic chemicals into the environment.     

Chlorination of drinking water and cancer incidence in Norway.

To examine whether chlorination of drinking water was associated with cancer of the digestive or other organs, an ecological epidemiological study using nationwide incidence data from the Cancer Registry of Norway was carried out. On two geographical levels (counties and municipalities), both for men and women, chlorination of drinking water was associated with an increased incidence of cancer of the colon and rectum. After adjusting for potential confounding variables, also measured on a geographical basis, the associations were still significant at the county level (adjusted for population density, income, education, fat and fibre intake etc.), but not at the municipality level. The observed associations are weak, chlorination being associated with a 20-40% increase in colorectal cancer rates. Due to inherent methodological limitations in ecological studies like the present one, causal interpretations should be made with great care. Thus, although the results give some support to the hypothesis that drinking water chlorination is associated with colorectal cancer, they do not provide strong evidence of a causal relationship.     

Residential radon and risk of lung cancer: a combined analysis of 7 North American case-control studies

BACKGROUND: Underground miners exposed to high levels of radon have an excess risk of lung cancer. Residential exposure to radon is at much lower levels, and the risk of lung cancer with residential exposure is less clear. We conducted a systematic analysis of pooled data from all North American residential radon studies. METHODS: The pooling project included original data from 7 North American case-control studies, all of which used long-term alpha-track detectors to assess residential radon concentrations. A total of 3662 cases and 4966 controls were retained for the analysis. We used conditional likelihood regression to estimate the excess risk of lung cancer. RESULTS: Odds ratios (ORs) for lung cancer increased with residential radon concentration. The estimated OR after exposure to radon at a concentration of 100 Bq/m3 in the exposure time window 5 to 30 years before the index date was 1.11 (95% confidence interval = 1.00-1.28). This estimate is compatible with the estimate of 1.12 (1.02-1.25) predicted by downward extrapolation of the miner data. There was no evidence of heterogeneity of radon effects across studies. There was no apparent heterogeneity in the association by sex, educational level, type of respondent (proxy or self), or cigarette smoking, although there was some evidence of a decreasing radon-associated lung cancer risk with age. Analyses restricted to subsets of the data with presumed more accurate radon dosimetry resulted in increased estimates of risk. CONCLUSIONS: These results provide direct evidence of an association between residential radon and lung cancer risk, a finding predicted using miner data and consistent with results from animal and in vitro studies.     

Cancer Mortality Rates in Appalachia: Descriptive Epidemiology and an Approach to Explaining Differences in Outcomes

Cancer is a leading cause of death in the Appalachian region of the United States. Existing studies compare regional mortality rates to those of the entire nation. We compare cancer mortality rates in Appalachia to those of the nation, with additional comparisons of Appalachian and non-Appalachian counties within the 13 states that contain the Appalachian region. Lung/bronchus, colorectal, female breast and cervical cancers, as well as all cancers combined, are included in analysis. Linear regression is used to identify independent associations between ecological socioeconomic and demographic variables and county-level cancer mortality outcomes. There is a pattern of high cancer mortality rates in the 13 states containing Appalachia compared to the rest of the United States. Mortality rate differences exist between Appalachian and non-Appalachian counties within the 13 states, but these are not consistent. Lung cancer is a major problem in Appalachia; most Appalachian counties within the 13 states have significantly higher mortality rates than in-state, non-Appalachian counterparts. Mortality rates from all cancers combined also appear to be worse overall within Appalachia, but part of this disparity is likely driven by lung cancer. Education and income are generally associated with cancer mortality, but differences in the strength and direction of these associations exist depending on location and cancer type. Improving high school graduation rates in Appalachia could result in a meaningful long term reduction in lung cancer mortality. The relative importance of household income level to cancer outcomes may be greater outside the Appalachian regions within these states.     

Testing a BEIR-VI suggestion for explaining the lung cancer vs. radon relationship for U.S. counties

The BEIR-VI Report suggests that the large discrepancy between the observed lung cancer rate vs. radon exposure relationship for U.S. counties, and the predictions of linear no-threshold theory, may be explained by a strong negative correlation between smoking intensity and radon exposure. It proposes a model for testing that suggestion. We apply that model to the detailed data for U.S. counties; analysis shows that even a perfect negative correlation explains little more than half of the discrepancy, and the largest not-implausible correlation can explain less than a quarter of the discrepancy. We then extend the BEIR-VI suggestion to include a strong negative correlation between both the prevalence of smoking and the intensity of smoking. The largest not-implausible correlations can explain no more than 30% of the discrepancy. It is concluded that the previous interpretation of these data, that linear no-threshold theory fails this test, is sustained.     

Gastrointestinal cancer mortality in New Jersey counties, and the relationship with environmental variables

The State of New Jersey (NJ) USA has been thought to have an unusually high cancer mortality rate; this assumption has been based on 1950-1969 mortality data for its 21 counties. This paper presents an analysis of gastrointestinal (GI) cancer mortality rates in New Jersey counties during 1968-1977, a comparison with the 1950-1969 rates, and associations between current GI cancer mortality rates and selected environmental variables. Age-adjusted mortality rates for GI cancers were calculated for the 21 NJ counties during the period 1968-1977, and were compared with the period 1950-1969, with the Surveillance, Epidemiology and End Results (SEER) survey and with cancer mortality in the US, 1973-1977. The county rates were also correlated with: the distribution of chemical toxic waste disposal sites; annual per capita income; the rates of low birth weight, birth defects, and infant mortality; chemical industry distribution; percentage of the population employed in chemical industries; the density of population; and the urbanization index for each of the counties. Some of the major findings are: Age-adjusted GI cancer mortality rates (all sites combined) were higher than national rates in 20 of 21 NJ counties. In comparison with national trends, NJ stomach cancer rates have declined less, oesophageal cancer rates have declined more, and pancreatic cancer mortality rates have followed similar patterns. Cancer mortality rates in NJ during the period 1968-1977 significantly (p less than 0.0001) exceeded national rates for cancer of the oesophagus (white male, non-white male), stomach (men and women), colon (white male, white female, non-white female), and rectum (whites only). In 18 of the 21 NJ counties, the observed number of cancer deaths for at least one GI cancer site was significantly greater than expected at the 0.0001 level for at least one population subgroup. Among white men, a significant (p less than 0.0001) excess of observed over expected cancer deaths was observed for three or more GI cancer sites in seven counties. The environmental variables that were most frequently associated with GI cancer mortality rates (except pancreatic cancer) were degree of urbanization, population density, and chemical toxic waste disposal sites. Some of the implications of the study findings are discussed and recommendations made for future investigations.     

Geographic patterns of pulmonary disease in south carolina

PURPOSE: The purpose of this study is to assess the patterns of geographic variation of the following pulmonary diseases: tuberculosis (TB), lung cancer (LC), berylliosis (BR), sarcoidosis (SR), and asthma (AS).METHODS: Hospitalization rates for the five diseases were calculated for the 46 South Carolina (SC) counties from 1985-1995. Variation in rates was assessed by county, region, and urban/rural status using analysis of variance. SC was divided into 4 regions, according to proximity to the coastline and to physiographic characteristics. Region 1 comprised counties in the western-most part of the state, and region 4 comprised the counties closest to the coastline. SC was also divided into 5 population density settings: extreme rural (<15,000), rural (15,000-30,000), middle-size (30,000-50,000), middle-urban (50,000-100,000) and urban (>100,000). Statistical analyses were performed with SAS Version 8.RESULTS:Table 1The rates of SR, TB, and BR were significantly higher in the coastal region. In contrast, no coastal pattern was detected by population density.CONCLUSIONS: The consistent association of SR, TB, and BR with geographical region, as opposed to the inconsistent association with population density, suggests that environmental factors may be involved in these pulmonary diseases. These results demonstrate the need for further analysis of the relationship between socio-economic, behavioral, and environmental factors and the development of these diseases.     

Cancer mortality in U.S. counties with hazardous waste sites and ground water pollution

Since the late 1950s, more than 750 million tons of toxic chemical wastes have been discarded in an estimated 30,000 to 50,000 hazardous waste sites (HWSs). Uncontrolled discarding of chemical wastes creates the potential for risks to human health. Utilizing the National Priorities Listing (NPL) of hazardous waste sites developed by the United States Environmental Protection Agency (EPA), this study identified 593 waste sites in 339 U.S. counties in 49 states with analytical evidence of contaminated ground drinking water providing a sole source water supply. For each identified county, age-adjusted, site-specific cancer mortality rates for 13 major sites for the decade 1970-1979, for white males and females, were extracted from U.S. Cancer Mortality and Trends 1950-1979. Also, HWS and non-HWS counties that showed excess numbers of deaths were enumerated for each cancer selected. Significant associations (p less than .002) between excess deaths and all HWS counties were shown for cancers of the lung, bladder, esophagus, stomach, large intestine, and rectum for white males; and for cancers of the lung, breast, bladder, stomach, large intestine, and rectum for white females when compared to all non-HWS counties. There were no consistent geographical patterns that suggested a broad distribution of gastrointestinal cancers associated with HWSs throughout the United States, although we did identify a cluster of excess gastrointestinal cancers in counties within states located in EPA Region 3 (Delaware, Maryland, Pennsylvania, Virginia, West Virginia).(ABSTRACT TRUNCATED AT 250 WORDS)