Mortality from cancers of the oral cavity and pharynx, oesphagus, larynx and lung between 1955 and 1989 has been analysed for USA, Canada and 14 countries in Latin America. Among males, Uruguay, Cuba, Argentina and Puerto Rico have the highest rates for all sites, and Peru, Ecuador, Dominican Republic, Mexico and Colombia have the lowest rates. Among females, Cuba, Colombia and Puerto Rico rank high for all sites, and Mexico, Paraguay, Ecuador and Peru rank low. For both sexes, lung cancer mortality rates from the US and Canada are high, whereas rates from other sites are intermediate. An increasing trend in lung cancer mortality over time is shown in all countries except Cuba (no changes), Argentina, Paraguay and Peru (decreasing trend). In Latin America, the tobacco-related lung cancer epidemic is in its early phase among males, and very early phase among females.

We computed four different correlation coefficients to investigate the degree of reproducibility of the weekly consumption of 77 food items or groups of foods and of seven summary questions from a food frequency questionnaire developed in Italy for a case-control study on cancers of the breast and digestive tract. The questionnaire had been administered twice to 452 Italian men and women. These included Pearson correlation coefficients (a) using the weekly frequencies of consumption without any transformation (P1); (b) after applying the transformation log (x + 1) (P2); (c) after applying the transformation log (x + 0.01) (P3); and (d) the Spearman correlation coefficient (SP). The mean values were 0.55 for P1, 0.59 for P2, 0.56 for P3 and 0.59 for SP. All coefficients were positively correlated, although to variable extents: the Spearman correlation coefficient between P2 and SP was 0.92, and that between P1 and P3 was 0.53. Differences between the four coefficients were more marked for food items with a lower kappa statistic and lower intraclass correlation, ie for those items with more severe reproducibility problems. Thus, a single correlation coefficient may not be enough to detect zones of the distribution of a food item where misclassification problems are more severe. The correlation coefficients used to investigate reproducibility should therefore be chosen on the basis of subsequent data analyses.

The relationship between diet diversity (i.e., variety in food intake computed as the total number of foods consumed at least once per week) and the risk of colorectal cancer was investigated using data from a hospital-based, case-control study carried out between 1985 and 1992 in Northern Italy. Subjects were patients with histologically confirmed incident cancers of the colon (n = 828) and rectum (n = 498) and 2024 controls admitted for acute, nonneoplastic, non-digestive tract conditions. A significant inverse association (multivariate relative risk, 0.7) for the highest (more diverse diet) versus the lowest quartile of total diet diversity was observed, together with a significant trend in risk. A similar risk was observed for diversity within vegetables (RR, 0.6, highest versus lowers quartile), but no consistent association was found for fruit, meat and fish, and "other food" diversity. A similar pattern was observed when colon and rectal cancers were considered separately or across separate strata of sex and age. Diet diversity is related to a moderately decreased risk of colorectal cancer. These results add epidemiological support to the dietary guidelines recommending a more varied diet, and, if confirmed by other studies, would have considerable public health implications.

We compared data from interviews of patients admitted to hospital and interviews repeated at home for 400 hospital controls. Of these, 294 (73.5%) could be re-interviewed (responders). Never smoking was reported by 44% of subjects in the in-hospital interviews and by 40% in the repeat at-home interviews. Among those interviewed in hospital, 38% claimed to be nondrinkers, as did 38% of responders, but only 28% characterized themselves as nondrinkers in the repeat interview. For consumption of regular coffee, 15% reported none in the selected sample, 14% among responders, but only 6% in the repeat at-home interview. Agreement was good for decaffeinated coffee and tea consumption.

<h3>To the Editor.</h3> —For decades, there has been a widespread social acceptance of cigarette smoking in Italy. The Ministry of Health (or any other governmental body) did not make or support any serious campaign to inform the public about the risks and damages of smoking, even when these were clear on a scientific level and most other Western countries took measures to control smoking.¹Thus, any manufacturer operating in the Italian market was allowed to sell cigarettes without information or warning labels,²and the Italian Parliament imposed some warning only in 1991—more than two decades after the United States, Canada, and most other Western countries.³ Mr Mario Stalteri was a pack-a-day smoker for 40 years, mostly of cigarettes produced by the state-owned manufacturer. In 1991, he died of lung cancer at 64 years of age, 4 years after stopping smoking. Following US developments in the topic of

The percent population attributable risk (AR) for kidney cancer was estimated in relation to smoking habits, beta-carotene intake, history of cystitis and family history of kidney cancer, using data from a case-control study conducted between 1985 and 1989 in Milan, northern Italy. The data comprised 133 histologically confirmed cases of incident kidney cancer and 392 controls, admitted to hospital for a wide range of acute, non-neoplastic, non-smoking-related diseases. On the basis of multivariate odds ratios (ORs), smoking habits accounted for about 26% of cases, a low beta-carotene intake for 18%, a history of cystitis for 7%, and a family history of kidney cancer in first-degree relatives for 3% of cases. Ever smoking and low beta-carotene intake combined explained 38% of all kidney cancers, and the combination of these two factors plus a history of cystitis and a family history of kidney cancer explained 45% of the incidence of the disease. Thus, even if available, dietary information was limited and the AR estimates were based on somewhat arbitrary assumptions. A considerable proportion of kidney cancers could be avoided simply by eliminating smoking and increasing consumption of fruit and vegetables in this Italian population. This would mean that about 1,500 kidney cancer deaths every year in the whole of Italy could be avoided.