Leukemia,Multiple Myeloma,and Solvent Exposure 2 Forli was essary in order to limit the study to all subjects for whom a detailed assessment of cases（anc l are thele higher th in the general rovince of Varese.where NHL that include CLL but not MM and leukemias were collected). The highest category was used for subjects who worked in Data Collection and coatolnesureeTo.enfyan Exposure Assessment of questionnaires.Furthermore,in order to reduce the All case and controls received a letter asking if they exposure assessment variability of the different experts,a y and ac interviewe job exposure matr is developed colle exposure assessment Data Analysis ad Data analyses were software [SAS,1999.P a job specific questionnaire ation extra and th 95%C and MM.AUl the diagnostic or therapeutic specific medications. logistic regression models,taking into account relevant family medical history.and reproductiv Ccppoaeh confounders (age.gender,education,and area) bjects who neve ste probability of exposure were excluded.Two collapse classes of exposure were considered:(D)very low and low questionnaire versus）medium ev to the the individual level for each case and control The blind with status categories.The linear test for trend used the mid-points of the wing categorie dered duration categones 5,and 35 years aliphatic hvdr RESULTS bons:and oxygenated derivative hydrocarbons.Evaluation for chemicals: The mair characteristics of the interviewed subjects are ene; nd MM and to chemica classes of solvents and individual agents are presented in Exposures were rated on two scales: “probability Table II.There was essentially no evidence for increased represented the nsk of all leuk emia for any solvent exposure.When looking the at sub-typ given calendar period.It was classified into three levels AML.For his mor the ors for expo sure to benzene were low,medium,high.The second scalew was "intensity, which 0.3(95%CI:0.1-1.0)for very low/low exposure and 0.9 ep sented the esti of t age in th 95%C: or m /high exposure (compare 263 cases of MM collected in 6 areas (Torino, Imperia, Forlı`, Firenze, Ragusa, Latina), and 1,100 controls sampled in the same areas. Restrictions were necessary in order to limit the study to all subjects for whom a detailed assessment of exposure to solvents had been performed, as described below. Results concerning CLL cases (and their controls) are reported from eight areas (the seven just indicated plus the province of Varese, where NHL, that include CLL, but not MM and leukemias were collected). Data Collection and Exposure Assessment All cases and controls received a letter asking if they agree to participate to the study and accept to be interviewed by trained personnel through a specific questionnaire. We conducted person-to-person interviews primarily at the interviewee’s home. Individuals affected by acute leukemia were interviewed mostly in the hospital. We obtained information on education, relevant life style factors such as tobacco smoking and beverage consumption, a detailed occupational history (including a collection of detailed data on specific jobs through the use of a job specific questionnaire for occupation [JSQ]), extra￾occupational exposure to solvents and pesticides, hair dye use, lifelong residential history, previous diseases, use of diagnostic or therapeutic X-rays, specific medications, family medical history, and reproductive history. The exposure assessment approach used in this study utilized job or industry-specific questionnaires and subse￾quent expert ratings in order to assign a level of exposure to a definitive list of agents. Industrial hygiene experts from each geographic area were selected to examine questionnaires and assess a level of probability and intensity of exposure to classes of solvents as well as substances at the individual level for each case and control. The assessment was blind with respect to case/control status. The following categories of solvents were considered: solvents in general; aromatic hydrocarbons; chlorinated hydrocarbons; technical hydrocarbons; aliphatic hydrocar￾bons; and oxygenated derivative hydrocarbons. Evaluation was also made for exposure to these specific chemicals: benzene; chloroform; dichloromethane; 1,4-dioxane; styr￾ene; tetrachloroethylene; trichloroethylene; 1,1,1-trichloro￾ethane; toluene; and xylene. Exposures were rated on two scales: ‘‘probability’’ represented the certainty/uncertainty of exposure based on knowledge of the materials used and technologies applied in the particular activity/production process reported in a given calendar period. It was classified into three levels: low, medium, high. The second scale was ‘‘intensity,’’ which represented the estimated concentration of the agent in the work environment and was measured on a 4-point scale: very low, low, medium, and high. The first class of intensity, ‘‘very low,’’ was used for subjects judged to have occupational exposure intensities comparable to the upper end of the normal range of exposures for the general population. ‘‘Low level’’ intensity was assigned when workplace exposure was judged to be low because of control measures, but never￾theless higher than in the general population. Medium exposure was assigned to subjects who worked where only moderate or poor control measures were judged to exist. The highest category was used for subjects who worked in workplaces with no control measures. To ensure a stand￾ardized approach, the assessors were centrally trained prior, and periodically during their independent evaluation of questionnaires. Furthermore, in order to reduce the exposure assessment variability of the different experts, a job exposure matrix was developed collecting the minimum overall consensus for the most frequent job titles/sectors in the study areas and it was used as baseline for the individual exposure assessment. Data Analysis Data analyses were performed with SAS statistical software [SAS, 1999]. Point estimates of odds ratios (OR) and the corresponding 95% confidence intervals (95% CI) were calculated. We computed ORs separately for leukemia and MM. All the analyses were performed using multiple logistic regression models, taking into account relevant potential confounders (age, gender, education, and area). Subjects who never used any of the listed chemicals were used as the referent population. Subjects assigned with a low probability of exposure were excluded. Two collapsed classes of exposure were considered: (I) very low and low versus (II) medium/high. Analyses for duration of exposure considered two levels: less than 15, and 15 or more years, and these were each compared to the never exposed referent group. Tests for trend were conducted across these three categories. The linear test for trend used the mid-points of the duration categories (0, 7.5, and 35 years). RESULTS The main characteristics of the interviewed subjects are reported in Table I. ORs for leukemia and MM and exposure to chemical classes of solvents and individual agents are presented in Table II. There was essentially no evidence for increased risk of all leukemia for any solvent exposure. When looking at sub-types of leukemia, we found no association between exposure to any chemical class or specific solvent and AML. For this tumor, the ORs for exposure to benzene were 0.3 (95% CI: 0.1–1.0) for very low/low exposure and 0.9 (95% CI: 0.4–2.3) for medium/high exposure (compared to the unexposed). Although confidence intervals were wide, there were elevated point estimates for the association Leukemia, Multiple Myeloma, and Solvent Exposure 3