Abstract

Background: There is some evidence that alcohol consumption is inversely associated with renal cell cancer (RCC), but the issue is still unclear.

Patients and methods: We investigated the relation using data from two Italian multicentric case–control studies conducted from 1985 to 2004, including a total of 1115 incident, histologically confirmed cases and 2582 controls hospitalised with acute, non-neoplastic conditions.

Results: Compared with non-drinkers, the multivariate odds ratios (ORs) of RCC were 0.87 [95% confidence interval (CI) 0.73–1.04] for ≤4 drinks per day, 0.76 (95% CI 0.59–0.99) for >4 to ≤8 drinks per day and 0.70 (95% CI 0.50–0.97) for >8 drinks per day of alcoholic beverages, with a significant inverse trend in risk (P value = 0.01). The ORs were 0.85 (95% CI 0.71–1.02) for wine, 0.84 (95% CI 0.68–1.03) for beer and 0.86 (95% CI 0.70–1.05) for spirits consumption, as compared with abstainers. No trend in risk of RCC emerged with duration (P value = 0.94) and age at starting alcohol consumption (P value = 0.81). Results were consistent in men and women, as well as in strata of age, smoking and body mass index.

Conclusions: This pooled analysis found an inverse association between alcohol drinking and RCC. Risks continued to decrease even above eight drinks per day (i.e. >100 g/day) of alcohol intake, with no apparent levelling in risk.

introduction

Epidemiological studies showed a role of a few modifiable risk factors in the aetiology of renal cell cancer (RCC), namely tobacco smoking and obesity [1]. Alcohol drinking has been less extensively investigated, but there is some evidence that it might be inversely (rather than directly, as for upper aerodigestive tract, liver, colorectal and breast cancer [2, 3]) associated with RCC. In fact, various cohort studies recently found a reduced incidence of RCC in alcohol drinkers, with relative risks (RRs) of 0.5 [95% confidence interval (CI) 0.3–0.8] in Finland [4] and 0.7 (95% CI 0.4–1.0) in the United States [5] and in Sweden (95% CI 0.4–1.2) [6], by comparing subjects at the highest level of alcohol consumption with non-drinkers or occasional drinkers. A pooled analysis of these and other eight cohort studies conducted in the United States, Canada and the Netherlands, including a total of 1430 subjects with RCC, showed RRs of 0.97, 0.82 and 0.72 for subsequent levels of alcohol consumption (highest level ≥15 g/day) as compared with non-drinkers and a similar inverse association in men and women [7].

Case–control studies generally found null or, more frequently in women, inverse associations [8–16]. The largest retrospective investigations providing information on alcohol and RCC were the ‘International renal cell cancer study’, including 1185 cases and 1526 controls that found odds ratios (ORs) for high alcohol intake of 1.0 (95% CI 0.7–1.4) in men and 0.5 (95% CI 0.3–0.8) in women [13], and a Canadian study of 1279 cases, which reported an inverse trend in risk of RCC in both genders (P value < 0.01) [17]. A few smaller case–control studies were conducted in Southern European populations [11, 15, 18, 19], where wine is the most commonly consumed alcoholic beverage [20], and these found no significant associations. The role of different types of alcoholic beverages, amount and duration of consumption and the consequences of drinking cessation are other aspects of the relation between alcohol consumption and RCC that are still unclear.

Therefore, we analysed the association of alcohol with RCC in Italy, i.e. in a population where regular and relatively heavy alcohol (mainly wine) drinking is common, using data from two case–control studies of RCC conducted from 1985 to 2004.

patients and methods

The general scheme and methods of the studies have been extensively described elsewhere [15, 21]. In the first study, data were collected from 1985 to 1992 in the greater Milan area and the province of Pordenone, in northern Italy, on 348 patients with RCC (236 men and 112 women) aged 25–77 years (median age 60 years) and 1048 controls (753 men and 295 women) aged 23–79 years (median age 60 years). The second study was conducted from 1992 to 2004 in four Italian areas, including the greater Milan area and the provinces of Udine and Pordenone in northern Italy, the province of Latina in central Italy and the urban area of Naples in southern Italy, on 767 patients with RCC (494 men and 273 women) aged 24–79 years (median age 62 years) and 1534 controls (988 men and 546 women) aged 22–79 years (median age 62 years), matched with cases by study centre, sex and quinquennia of age, with a case to control ratio of 1 : 2. In both studies, cases were patients with incident, histologically confirmed RCC, admitted to major teaching and general hospitals of the study areas. Cancers of the renal pelvis and ureter (International Classification of Diseases-9 189.1–189.2) were not included. Histological information from the second study included cancer cell type [22]. There were 416 clear-cell, 51 papillary and 11 chromophobic carcinomas, 88 other/unclassified subtypes and 201 unknown histology. Controls were admitted to the same hospitals as cases for a wide spectrum of acute, non-neoplastic conditions, unrelated to known risk factors for RCC. The distribution of control diagnoses was different in the two studies; overall, 24% of controls were admitted for traumas (mostly fractures and sprains), 30% for other orthopaedic disorders (such as low-back pain and disc disorders), 17% for surgical conditions and 29% for various other illnesses including eye, nose, ear or skin disorders. Less than 5% of both cases and controls contacted refused to participate. A total of 1115 cases and 2582 controls were included in the present analysis.

Cases and controls were interviewed during their hospital stay by trained interviewers, using structured questionnaires that included information on socio-demographic characteristics; anthropometric measures; lifestyle habits, such as tobacco smoking and coffee consumption; dietary habits; personal medical history and family history of cancer in first-degree relatives. The section on alcohol drinking of the first study included the number of days per week and drinks per day each alcoholic beverage (i.e. wine, beer and spirits) was consumed and the duration of the habit. The second study included questions on weekly number of drinks of wine, beer, grappa (a typical Italian spirit), amari and digestives (other types of Italian liquors drunk after meals) and spirits (whisky, cognac and brandy), age at first and last use of each alcoholic beverage and whether consumption occurred during or outside meals or both. In both studies, one drink corresponded ∼ 125 ml of wine, 330 ml of beer and 30 ml of liquors and spirits (i.e. ∼12–15 g of alcohol, on the basis of a specifically designed food and beverages composition database [23]). The period of drinking was investigated up to 1 year before diagnosis for cases or hospital admission for controls. The questionnaires were tested for reproducibility and validity, and a satisfactory level of both measures of the pattern of alcohol consumption was observed across different levels and types of alcoholic beverage intake [24, 25]. The reproducibility of wine and total alcohol intake showed r > 0.75, whereas the validity was somewhat higher for wine (r ∼ 0.70) than for other alcoholic beverages and total alcohol.

We used a two-sided Wilcoxon rank sum test to compare cases and controls according to consumption of different types of alcoholic beverages. ORs and the corresponding 95% CIs were calculated using multiple logistic regression models [26], adjusted for study, sex, age, study centre, years of education, family history of renal cancer in first-degree relatives, tobacco smoking and body mass index (BMI). The OR for an increase of one drink per day was also estimated, including total alcohol (or wine) consumption in the model as a continuous variable. The dose–response relation between ethanol daily intake and risk of RCC was further investigated using logistic regression splines with appropriate pointwise CIs [27–29], adjusted for the same cofactors. The best-fitting model was selected minimising the Akaike information criterion [30].

results

The distribution of RCC cases and of the comparison group according to selected variables is presented in Table 1. As compared with controls, cases reported a higher educational level (P < 0.01), were more frequently heavy smokers (P < 0.05) and had more frequently a family history of kidney cancer in first-degree relatives (P < 0.01). Among drinkers of each type of alcoholic beverage, the mean numbers of daily drinks were 2.95 and 3.32 for cases and controls, respectively, for wine consumption (P < 0.01), 0.67 and 1.16 for beer consumption (P < 0.01) and 0.73 and 0.84 for spirits consumption (P < 0.05).

Table 1.

Distribution of 1115 cases of renal cell carcinoma and 2582 controls according to sex, age and selected covariates (Italy, 1985–2004)

 Cases (%) Controls (%) χ2 
Sex 
    Males 730 (65) 1741 (67) 1.34 
    Females 385 (35) 841 (33) 
Age (years) 
    <50 185 (17) 428 (17) 0.71 
    50–59 311 (28) 726 (28) 
    60–69 417 (37) 935 (36) 
    70–79 202 (18) 493 (19) 
Education (years) 
    <7 579 (52) 1542 (60) 31.05** 
    7–11 302 (27) 675 (26) 
    ≥12 234 (21) 365 (14) 
Smoking habita 
    Never smokers 452 (41) 1030 (40) 9.98* 
    Ex-smokers 285 (25) 681 (26) 
    Current smokers   
    1–19 cigarettes per day 178 (16) 497 (19) 
    ≥20 cigarettes per day 196 (18) 372 (15) 
Family history of kidney cancerb 
    No 1092 (98) 2566 (99) 15.53** 
    Yes 23 (2) 16 (1) 
Type of beverage (drinks per day)c 
    Wine 2.95 ± 2.5 3.32 ± 2.8 P = 0.007 
    Beer 0.67 ± 1.5 1.16 ± 2.8 P < 0.001 
    Spirits 0.73 ± 1.1 0.84 ± 1.5 P = 0.04 
 Cases (%) Controls (%) χ2 
Sex 
    Males 730 (65) 1741 (67) 1.34 
    Females 385 (35) 841 (33) 
Age (years) 
    <50 185 (17) 428 (17) 0.71 
    50–59 311 (28) 726 (28) 
    60–69 417 (37) 935 (36) 
    70–79 202 (18) 493 (19) 
Education (years) 
    <7 579 (52) 1542 (60) 31.05** 
    7–11 302 (27) 675 (26) 
    ≥12 234 (21) 365 (14) 
Smoking habita 
    Never smokers 452 (41) 1030 (40) 9.98* 
    Ex-smokers 285 (25) 681 (26) 
    Current smokers   
    1–19 cigarettes per day 178 (16) 497 (19) 
    ≥20 cigarettes per day 196 (18) 372 (15) 
Family history of kidney cancerb 
    No 1092 (98) 2566 (99) 15.53** 
    Yes 23 (2) 16 (1) 
Type of beverage (drinks per day)c 
    Wine 2.95 ± 2.5 3.32 ± 2.8 P = 0.007 
    Beer 0.67 ± 1.5 1.16 ± 2.8 P < 0.001 
    Spirits 0.73 ± 1.1 0.84 ± 1.5 P = 0.04 
a

The sum does not add up to the total because of some missing values.

b

In first-degree relatives.

c

Mean ± standard deviation, calculated among drinkers of each alcoholic beverage; P values were calculated by a two-sided Wilcoxon rank sum test.

*P < 0.05; **P < 0.001.

Table 2 gives the frequency distribution of cases and controls, the multivariate ORs and corresponding 95% CIs according to various measures of alcohol consumption. Overall, 23.1% of cases and 19.2% of controls were non-drinkers, whereas 6.3% of cases and 9.2% of controls consumed more than eight drinks of alcoholic beverages per day. With reference to the distribution of consumption of different types of beverages, 75.4% of cases and 78.8% of controls reported wine consumption, 32.8% of cases and 33.3% of controls reported beer consumption and 33.2% of cases and 34.3% of controls reported spirits consumption. As compared with non-drinkers, the ORs were 0.87 (95% CI 0.73–1.04) for subjects consuming ≤4 drinks of alcoholic beverages per day, 0.76 (95% CI 0.59–0.99) for >4 to ≤8 drinks per day and 0.70 (95% CI 0.50–0.97) for >8 drinks per day. There was a significant inverse trend in risk with dose (P value = 0.01). The OR was 0.97 (95% CI 0.95–1.00) for an increase in consumption of one drink of alcoholic beverages per day. With reference to different alcoholic beverages, ORs were 0.85 (95% CI 0.71–1.02) for wine, 0.84 (95% CI 0.68–1.03) for beer and 0.86 (95% CI 0.70–1.05) for spirits, as compared with abstainers. When we examined the frequency of days per week of wine consumption, the ORs were 0.97 (95% CI 0.73–1.28) for discontinuous drinkers and 0.83 (95% CI 0.69–1.00) for regular daily drinkers, as compared with abstainers. Further adjustment of the latter estimates for amount of alcohol consumption did not materially change the results. There was no trend in risk of RCC with duration of the habit (P value = 0.94) and age at starting consumption (P value = 0.81), nor any difference emerged comparing ex-drinkers (63 cases and 114 controls) with never drinkers (OR = 0.98, 95% CI 0.68–1.41).

Table 2.

Distribution of 1115 cases of RCC and 2582 controls according to alcohol consumption (Italy, 1985–2004)

 Cases Controls ORa (95% CI) 
 No No  
Total alcohol, drinks per day 
    Non-drinkers 258 23.1 495 19.2 1b 
    Current drinkers 857 76.9 2086 80.8 0.84 (0.71–1.01) 
        ≤4 617 55.3 1371 53.1 0.87 (0.73–1.04) 
        >4 to ≤8 170 15.2 477 18.5 0.76 (0.59–0.99) 
        >8 70 6.4 238 9.2 0.70 (0.50–0.97) 
        P for trend     0.01 
    Continuous ORac     0.97 (0.95–1.00) 
Type of beverage 
    Wine drinkers 841 75.4 2035 78.8 0.85 (0.71–1.02) 
    Wine drinks per dayd 
        0–4 694 62.2 1623 62.9 0.85 (0.71–1.01) 
        >4 to ≤8 149 13.4 392 15.2 0.89 (0.68–1.16) 
        >8 14 1.3 71 2.7 0.50 (0.27–0.92) 
        P for trend     0.06 
    Continuous ORac     0.98 (0.95–1.00) 
    Wine, frequency of consumption 
        <7 days/week 117 10.5 219 8.5 0.97 (0.73–1.28) 
        Daily 724 65.0 1817 70.3 0.83 (0.69–1.00) 
    Beer drinkers 366 32.8 859 33.3 0.84 (0.68–1.03) 
    Spirits drinkers 370 33.2 885 34.3 0.86 (0.70–1.05) 
    Duration of drinking, years 
        ≤34 288 27.0 797 32.2 0.73 (0.58–0.91) 
        35–44 288 27.0 655 26.4 0.91 (0.73–1.14) 
        ≥45 233 21.8 532 21.5 0.97 (0.76–1.24) 
        P for trend     0.94 
    Age at startinge, years 
        ≤16 129 24.4 271 24.8 1b 
        17–19 92 17.4 198 18.1 0.94 (0.67–1.32) 
        20–22 162 30.7 331 30.3 0.99 (0.74–1.33) 
        ≥23 145 27.5 292 26.7 1.02 (0.75–1.39) 
        P for trend     0.81 
 Cases Controls ORa (95% CI) 
 No No  
Total alcohol, drinks per day 
    Non-drinkers 258 23.1 495 19.2 1b 
    Current drinkers 857 76.9 2086 80.8 0.84 (0.71–1.01) 
        ≤4 617 55.3 1371 53.1 0.87 (0.73–1.04) 
        >4 to ≤8 170 15.2 477 18.5 0.76 (0.59–0.99) 
        >8 70 6.4 238 9.2 0.70 (0.50–0.97) 
        P for trend     0.01 
    Continuous ORac     0.97 (0.95–1.00) 
Type of beverage 
    Wine drinkers 841 75.4 2035 78.8 0.85 (0.71–1.02) 
    Wine drinks per dayd 
        0–4 694 62.2 1623 62.9 0.85 (0.71–1.01) 
        >4 to ≤8 149 13.4 392 15.2 0.89 (0.68–1.16) 
        >8 14 1.3 71 2.7 0.50 (0.27–0.92) 
        P for trend     0.06 
    Continuous ORac     0.98 (0.95–1.00) 
    Wine, frequency of consumption 
        <7 days/week 117 10.5 219 8.5 0.97 (0.73–1.28) 
        Daily 724 65.0 1817 70.3 0.83 (0.69–1.00) 
    Beer drinkers 366 32.8 859 33.3 0.84 (0.68–1.03) 
    Spirits drinkers 370 33.2 885 34.3 0.86 (0.70–1.05) 
    Duration of drinking, years 
        ≤34 288 27.0 797 32.2 0.73 (0.58–0.91) 
        35–44 288 27.0 655 26.4 0.91 (0.73–1.14) 
        ≥45 233 21.8 532 21.5 0.97 (0.76–1.24) 
        P for trend     0.94 
    Age at startinge, years 
        ≤16 129 24.4 271 24.8 1b 
        17–19 92 17.4 198 18.1 0.94 (0.67–1.32) 
        20–22 162 30.7 331 30.3 0.99 (0.74–1.33) 
        ≥23 145 27.5 292 26.7 1.02 (0.75–1.39) 
        P for trend     0.81 

The sums may not add up to the total because of some missing values.

a

Estimates from multiple logistic regression including terms for age, sex, study, study centre, education, smoking habits, body mass index and family history of renal cancer.

b

Reference category. Alcohol abstainers are the reference category also for analyses on type of beverage, frequency of consumption and duration of drinking.

c

For an increment of one drink per day.

d

The sum does not add up to the total number of wine drinkers because drinkers of alcoholic beverages other than wine (16 cases and 51 controls) were included in the category ‘0–4 drinks per day’.

e

Data from the second study only (among 767 cases and 1534 controls from 1992 to 2004).

OR, odds ratio; CI, confidence interval.

Table 3 gives the ORs for total alcohol consumption according to strata of selected covariates. Results were consistent for men and women (P value for heterogeneity, 0.67). Likewise, no significant heterogeneity was reported across strata of age (P value = 0.92), BMI (P value = 0.40) and smoking (P value = 0.51). The risk in drinkers versus non-drinkers was, if anything, lower for the subset of 416 cases classified as clear-cell carcinomas (OR = 0.80, 95% CI 0.61–1.04).

Table 3.

Relationship between alcohol consumption and risk of renal cell cancer in strata of selected covariates (Italy, 1985–2004)

 Non-drinkers <3 drinks per day ≥3 drinks per day 
Sex 
    Males 
        Cases : controls 104 : 214 262 : 552 364 : 974 
        ORa (95% CI) 1b 0.92 (0.70–1.22) 0.89 (0.68–1.17) 
    Females 
        Cases : controls 154 : 281 200 : 475 31 : 85 
        ORa (95% CI) 1b 0.79 (0.61–1.03) 0.79 (0.49–1.27) 
    P for heterogeneity 0.67 
Age, years 
    <60 
        Cases : controls 116 : 219 204 : 444 176 : 490 
        ORa (95% CI) 1b 0.86 (0.64–1.15) 0.81 (0.58–1.11) 
    ≥60 
        Cases : controls 142 : 276 258 : 583 219 : 569 
        ORa (95% CI) 1b 0.85 (0.65–1.09) 0.85 (0.63–1.14) 
    P for heterogeneity 0.92 
Body mass index 
    <Median valuec 
        Cases : controls 120 : 259 252 : 535 184 : 498 
        ORa (95% CI) 1b 1.00 (0.77–1.31) 0.83 (0.61–1.12) 
    ≥Median value 
        Cases : controls 138 : 236 210 : 492 211 : 561 
        ORa (95% CI) 1b 0.74 (0.56–0.97) 0.85 (0.62–1.15) 
    P for heterogeneity 0.40 
Smoking habit 
    Never smokers 
        Cases : controls 151 : 270 212 : 492 89 : 267 
        ORa (95% CI) 1b 0.78 (0.60–1.01) 0.69 (0.48–0.99) 
    Ex-smokers 
        Cases : controls 45 : 89 107 : 244 133 : 348 
        ORa (95% CI) 1b 0.85 (0.55–1.31) 1.04 (0.67–1.63) 
    Current smokers 
        Cases : controls 62 : 136 141 : 290 171 : 443 
        ORa (95% CI) 1b 0.97 (0.67–1.41) 0.84 (0.57–1.23) 
    P for heterogeneity 0.51 
 Non-drinkers <3 drinks per day ≥3 drinks per day 
Sex 
    Males 
        Cases : controls 104 : 214 262 : 552 364 : 974 
        ORa (95% CI) 1b 0.92 (0.70–1.22) 0.89 (0.68–1.17) 
    Females 
        Cases : controls 154 : 281 200 : 475 31 : 85 
        ORa (95% CI) 1b 0.79 (0.61–1.03) 0.79 (0.49–1.27) 
    P for heterogeneity 0.67 
Age, years 
    <60 
        Cases : controls 116 : 219 204 : 444 176 : 490 
        ORa (95% CI) 1b 0.86 (0.64–1.15) 0.81 (0.58–1.11) 
    ≥60 
        Cases : controls 142 : 276 258 : 583 219 : 569 
        ORa (95% CI) 1b 0.85 (0.65–1.09) 0.85 (0.63–1.14) 
    P for heterogeneity 0.92 
Body mass index 
    <Median valuec 
        Cases : controls 120 : 259 252 : 535 184 : 498 
        ORa (95% CI) 1b 1.00 (0.77–1.31) 0.83 (0.61–1.12) 
    ≥Median value 
        Cases : controls 138 : 236 210 : 492 211 : 561 
        ORa (95% CI) 1b 0.74 (0.56–0.97) 0.85 (0.62–1.15) 
    P for heterogeneity 0.40 
Smoking habit 
    Never smokers 
        Cases : controls 151 : 270 212 : 492 89 : 267 
        ORa (95% CI) 1b 0.78 (0.60–1.01) 0.69 (0.48–0.99) 
    Ex-smokers 
        Cases : controls 45 : 89 107 : 244 133 : 348 
        ORa (95% CI) 1b 0.85 (0.55–1.31) 1.04 (0.67–1.63) 
    Current smokers 
        Cases : controls 62 : 136 141 : 290 171 : 443 
        ORa (95% CI) 1b 0.97 (0.67–1.41) 0.84 (0.57–1.23) 
    P for heterogeneity 0.51 

The sums may not add up to the total because of some missing values.

a

Estimates from multiple logistic regression including terms for age, sex, study, study centre, education, smoking habits, body mass index and family history of kidney cancer.

b

Reference category.

c

The median value was calculated among the distribution of controls and was equal to 25.9 kg/m2.

OR, odds ratio; CI, confidence interval.

Figure 1 shows the shape of the best-fitting regression spline (dose–response curve). The risk of RCC decreased throughout the range of alcohol consumption.

Figure 1.

Estimates of odds ratios (solid line) and 95% confidence intervals (dashed lines) of renal cell cancers using regression spline by total alcohol intake. Regression equations include terms for age, sex, study, study centre, education, smoking habits, body mass index and family history of renal cancer. Curves are shown for best-fitting splines according to Akaike information criterion.

Figure 1.

Estimates of odds ratios (solid line) and 95% confidence intervals (dashed lines) of renal cell cancers using regression spline by total alcohol intake. Regression equations include terms for age, sex, study, study centre, education, smoking habits, body mass index and family history of renal cancer. Curves are shown for best-fitting splines according to Akaike information criterion.

discussion

This uniquely large national investigation from a southern European population found an inverse association between alcohol drinking and RCC. Thus, this study confirmed the results of a recent pooled analysis of cohort studies [7]. Given the wide range of alcohol drinking in Italy, including a relatively common heavy wine consumption [31], this study, however, also allowed to evaluate the risk of RCC in heavy drinkers. We found that risks continued to decrease even above eight drinks of alcoholic beverages per day (i.e. >100 g/day of alcohol), with no apparent levelling in risk. The existence of an inverse dose–risk relation gave further support to the hypothesis of a protective effect of alcoholic beverages. The reduction in risk of RCC was consistent in men and women, as well as in strata of age, smoking habit and BMI.

Lee et al. [7] explained the inverse association between alcohol consumption and RCC through the effects of alcohol on insulin sensitivity [32]. Other possible mechanisms of protection are a role of fluid intake (by diluting carcinogenic substances) and/or of antioxidant compounds contained in alcoholic beverages. Nevertheless, we found no meaningful difference in risk among different alcoholic beverages.

Duration of the habit was not associated with RCC risk. Duration of alcohol consumption, however, showed no influence on cancer even in studies of the upper aerodigestive tract, i.e. sites that are strongly (directly) related with amount of alcohol intake [33–35]. In the second study only, we were also able to consider age at starting consumption of alcohol that showed no influence on the risk of developing RCC, too, and the effect of stopping drinking. Nevertheless, when we separated ex-drinkers from never drinkers, we found no difference in risk of RCC.

A few previous studies indicated a different effect of alcohol intake on RCC risk according to gender, with a protective outcome among women but not men [12–14]. We also had considered this issue in an earlier report using data from our first case–control study [15], and we did not find significant results in both men and women. This pooled analysis conducted on a larger number of subjects also showed no appreciable modification in risk by gender.

Hospital controls could differ from the general population with regard to alcohol consumption, but we carefully excluded from the study any patient admitted to the hospital for chronic conditions or digestive tract diseases. Underreporting of drinking may have occurred, but alcohol consumption is socially accepted in Italy and information on alcohol drinking from the questionnaire was satisfactorily reliable and valid [24, 25]. Further, the similar interview setting of cases and controls is reassuring against potential information bias. Among other strengths of this study are the almost complete participation of cases and controls, use of data from a population with a wide range of alcohol drinking and a peculiar pattern of consumption (i.e. mainly wine drinking, consumed regularly during meals by both men and women), availability of information on duration of alcohol drinking, of several potential covariates of alcohol consumption and of age at starting and stopping consumption of each beverage.

funding

Italian Association for Cancer Research; the Italian League against Cancer; the Italian Ministry of Education (PRIN 2005); Italian Foundation for Cancer Research to C.G.

The authors thank Mrs Ivana Garimoldi for editorial assistance.

References

1.
McLaughlin
JK
Lipworth
L
Tarone
RE
Blot
WJ
Schottenfeld
D
Fraumeni
JF
Renal cancer
Cancer Epidemiology and Prevention
 , 
2006
3rd edition
New York, NY
Oxford University Press
(pg. 
1087
-
1100
)
2.
International Agency for Research on Cancer
Monographs on the Evaluation of Carcinogenic Risks to Humans. Volume 44. Alcohol Drinking
 , 
1988
Lyon, France
IARC
3.
MacDonald
I
Health Issues Related to Alcohol Consumption
 , 
1999
2nd edition
Brussels, Belgium
ILSI Europe
4.
Mahabir
S
Leitzmann
MF
Virtanen
MJ
, et al.  . 
Prospective study of alcohol drinking and renal cell cancer risk in a cohort of Finnish male smokers
Cancer Epidemiol Biomarkers Prev
 , 
2005
, vol. 
14
 (pg. 
170
-
175
)
5.
Lee
JE
Giovannucci
E
Smith-Warner
SA
, et al.  . 
Total fluid intake and use of individual beverages and risk of renal cell cancer in two large cohorts
Cancer Epidemiol Biomarkers Prev
 , 
2006
, vol. 
15
 (pg. 
1204
-
1211
)
6.
Rashidkhani
B
Akesson
A
Lindblad
P
Wolk
A
Alcohol consumption and risk of renal cell carcinoma: a prospective study of Swedish women
Int J Cancer
 , 
2005
, vol. 
117
 (pg. 
848
-
853
)
7.
Lee
JE
Hunter
DJ
Spiegelman
D
, et al.  . 
Alcohol intake and renal cell cancer in a pooled analysis of 12 prospective studies
J Natl Cancer Inst
 , 
2007
, vol. 
99
 (pg. 
801
-
810
)
8.
Goodman
MT
Morgenstern
H
Wynder
EL
A case-control study of factors affecting the development of renal cell cancer
Am J Epidemiol
 , 
1986
, vol. 
124
 (pg. 
926
-
941
)
9.
Asal
NR
Risser
DR
Kadamani
S
, et al.  . 
Risk factors in renal cell carcinoma: I. Methodology, demographics, tobacco, beverage use, and obesity
Cancer Detect Prev
 , 
1988
, vol. 
11
 (pg. 
359
-
377
)
10.
Brownson
RC
A case-control study of renal cell carcinoma in relation to occupation, smoking, and alcohol consumption
Arch Environ Health
 , 
1988
, vol. 
43
 (pg. 
238
-
241
)
11.
Talamini
R
Baron
AE
Barra
S
, et al.  . 
A case-control study of risk factor for renal cell cancer in northern Italy
Cancer Causes Control
 , 
1990
, vol. 
1
 (pg. 
125
-
131
)
12.
Mellemgaard
A
Engholm
G
McLaughlin
JK
Olsen
JH
Risk factors for renal cell carcinoma in Denmark. I. Role of socioeconomic status, tobacco use, beverages, and family history
Cancer Causes Control
 , 
1994
, vol. 
5
 (pg. 
105
-
113
)
13.
Wolk
A
Gridley
G
Niwa
S
, et al.  . 
International renal cell cancer study. VII. Role of diet
Int J Cancer
 , 
1996
, vol. 
65
 (pg. 
67
-
73
)
14.
Parker
AS
Cerhan
JR
Lynch
CF
, et al.  . 
Gender, alcohol consumption, and renal cell carcinoma
Am J Epidemiol
 , 
2002
, vol. 
155
 (pg. 
455
-
462
)
15.
Pelucchi
C
La Vecchia
C
Negri
E
, et al.  . 
Alcohol drinking and renal cell carcinoma in women and men
Eur J Cancer Prev
 , 
2002
, vol. 
11
 (pg. 
543
-
545
)
16.
Greving
JP
Lee
JE
Wolk
A
, et al.  . 
Alcoholic beverages and risk of renal cell cancer
Br J Cancer
 , 
2007
, vol. 
97
 (pg. 
429
-
433
)
17.
Hu
J
Mao
Y
White
K
Diet and vitamin or mineral supplements and risk of renal cell carcinoma in Canada
Cancer Causes Control
 , 
2003
, vol. 
14
 (pg. 
705
-
714
)
18.
Mattioli
S
Truffelli
D
Baldasseroni
A
, et al.  . 
Occupational risk factors for renal cell cancer: a case–control study in northern Italy
J Occup Environ Med
 , 
2002
, vol. 
44
 (pg. 
1028
-
1036
)
19.
Benhamou
S
Lenfant
MH
Ory-Paoletti
C
Flamant
R
Risk factors for renal-cell carcinoma in a French case-control study
Int J Cancer
 , 
1993
, vol. 
55
 (pg. 
32
-
36
)
20.
WHO/Europe
European Health for All Database (HFA-DB)
  
World Health Organization Regional Office for Europe. 2007; http://www.euro.who.int/hfadb, (31 December 2007, date last accessed)
21.
Bravi
F
Bosetti
C
Scotti
L
, et al.  . 
Food groups and renal cell carcinoma: a case-control study from Italy
Int J Cancer
 , 
2006
, vol. 
120
 (pg. 
681
-
685
)
22.
Eble
JN
Sauter
G
Epstein
JI
Sesterhenn
IA
Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs
 , 
2004
Lyon, France
IARC Press
23.
Salvini
S
Parpinel
MT
Gnagnarella
P
, et al.  . 
Banca dati di composizione degli alimenti per studi epidemiologici in Italia
 , 
1998
Milan, Italy
Istituto Europeo di Oncologia
24.
Ferraroni
M
Decarli
A
Franceschi
S
, et al.  . 
Validity and reproducibility of alcohol consumption in Italy
Int J Epidemiol
 , 
1996
, vol. 
25
 (pg. 
775
-
782
)
25.
D'Avanzo
B
La Vecchia
C
Katsouyanni
K
, et al.  . 
Reliability of information on cigarette smoking and beverage consumption provided by hospital controls
Epidemiology
 , 
1996
, vol. 
7
 (pg. 
312
-
315
)
26.
Breslow
NE
Day
NE
Statistical Methods in Cancer Research. Vol 1: the Analysis of Case-Control Studies
 , 
1980
Geneva, Switzerland
IARC Science Publication 32
27.
Hastie
TJ
Tibshirani
RJ
Generalized Additive Models
 , 
1994
London, UK
Chapman & Hall
28.
Rosenberg
PS
Katki
H
Swanson
CA
, et al.  . 
Quantifying epidemiologic risk factors using non-parametric regression: model selection remains the greatest challenge
Stat Med
 , 
2003
, vol. 
22
 (pg. 
3369
-
3381
)
29.
Greenland
S
Dose-response and trend analysis in epidemiology: alternatives to categorical analysis
Epidemiology
 , 
1995
, vol. 
6
 (pg. 
356
-
365
)
30.
Akaike
H
Petrov
BN
Csàki
F
Information theory and an extension of the maximum likelihood principle
Second International Symposium on Information Theory
 , 
1973
Budapest, Hungary
Académiai Kiadò
(pg. 
267
-
281
)
31.
Italian National Institute of Statistics
Stili di vita e condizioni di salute. Indagine multiscopo sulle famiglie “Aspetti della vita quotidiana”
 , 
2005
Rome, Italy
ISTAT
32.
Kiechl
S
Willeit
J
Poewe
W
, et al.  . 
Insulin sensitivity and regular alcohol consumption: large, prospective, cross sectional population study (Bruneck study)
BMJ
 , 
1996
, vol. 
313
 (pg. 
1040
-
1044
)
33.
Franceschi
S
Talamini
R
Barra
S
, et al.  . 
Smoking and drinking in relation to cancers of the oral cavity, pharynx, larynx, and esophagus in northern Italy
Cancer Res
 , 
1990
, vol. 
50
 (pg. 
6502
-
6507
)
34.
Castellsague
X
Munoz
N
De Stefani
E
, et al.  . 
Independent and joint effects of tobacco smoking and alcohol drinking on the risk of esophageal cancer in men and women
Int J Cancer
 , 
1999
, vol. 
82
 (pg. 
657
-
664
)
35.
Franceschi
S
Levi
F
Dal Maso
L
, et al.  . 
Cessation of alcohol drinking and risk of cancer of the oral cavity and pharynx
Int J Cancer
 , 
2000
, vol. 
85
 (pg. 
787
-
790
)