Mortality risk in the French cohort of uranium miners: extended follow-up 1946-1999.

OBJECTIVES: This paper presents the risk of death from lung cancer and from other causes of death for the French cohort of uranium miners through 1999 and estimates associations with radon exposure. METHODS: The cohort includes men employed as uranium miners for at least 1 year between 1946 and 1990. For each miner, vital status and cause of death were obtained from the national registry, and radon exposure was reconstructed for each year. Standardised mortality ratios were computed with national mortality rates as references. Exposure-risk relationships were estimated by Poisson regression, with a linear excess relative risk (ERR) model and a 5-year lag. RESULTS: The cohort included 5086 miners and 153 063 person-years of exposure. The mean duration of follow-up was 30.1 years. In all 4140 miners exposed to radon, the average cumulative exposure was 36.6 working level months (WLM). There were 1411 deaths of miners <85 years of age. The miners did not differ significantly in overall mortality from the general male population. The analysis confirmed an excess risk of lung cancer death (n = 159; SMR = 1.43; 95% CI: 1.22 to 1.68), which increased significantly with cumulative radon exposure (ERR per 100 WLM = 0.71; 95% CI: 0.29 to 1.35). The ERR per unit exposure was much higher after 1955, when the accuracy of exposure measurement improved substantially (ERR per 100 WLM = 2.00; 95% CI: 0.91 to 3.65). A significant excess of kidney cancer deaths was observed (n = 20; SMR = 2.0; 95% CI: 1.22 to 3.09), which was not associated with cumulative radon exposure. No excess was observed for other causes of death, except silicosis (n = 23; SMR = 7.12; 95% CI: 4.51 to 10.69). CONCLUSIONS: The analysis confirmed the excess risk of death from lung cancer associated with low radon exposure. An excess risk of death from kidney cancer was also observed, apparently not associated with cumulative radon exposure.

Indoor airborne endotoxin assessment in homes of Paris newborn babies.

The aims of this study were first to assess airborne endotoxin levels in the dwellings of 162 newborns living in Paris twice during a 1-year period, and second, to identify predictors for endotoxin concentrations using questionnaire data in relation to housing factors and living conditions. Air samples were collected on a glass fiber filter in polystyrene filter holders, using a pump at a flow rate of 3.5 l/min for 24 h placed in the main room of the home. Endotoxin levels were measured using a chromogenic kinetic Limulus Amoebocyte Lysate test. Geometric means (geometric standard deviation) of airborne endotoxin levels at two different visits were respectively 0.509 (4.289) EU/m3 and 0.557 (3.029) EU/m3. Airborne endotoxin levels were significantly increased: (i) in cold season (P = 0.024), with (ii) the presence of visible cockroaches in the previous 12 months at home (P < 0.001), (iii) increased number of inhabitants per square meter (P = 0.012), (iv) the high frequency of cleaning with the floor cloths (P = 0.0014), and (v) the low frequency of vacuuming (P = 0.0045). This study provided for the first time airborne endotoxin levels issued from repeated measurements in Paris dwellings. PRACTICAL IMPLICATIONS This analysis contributed to identify a few factors that determined indoor airborne endotoxin levels. However, the predictive model including housing factors and living conditions poorly estimated endotoxin levels. Consequently, multiple samples and longer sampling periods might improve the estimate of long-term airborne endotoxin exposure especially its variability, in cohort studies.