False-positive results are inherent in the scientific process of testing hypotheses concerning the determinants of cancer and other human illnesses. Although much of what is known about the etiology of human cancers has arisen from well-conducted epidemiological studies, epidemiology has been increasingly criticized for producing findings that are often sensationalized in the media and fail to be upheld in subsequent studies. Herein we describe examples from cancer epidemiology of likely false-positive findings and discuss conditions under which such results may occur. We suggest general guidelines or principles, including the endorsement of editorial policies requiring the prominent listing of study caveats, which may help reduce the reporting of misleading results. Increased epistemological humility regarding findings in epidemiology would go a long way to diminishing the detrimental effects of false-positive results on the allocation of limited research resources, on the advancement of knowledge of the causes and prevention of cancer, and on the scientific reputation of epidemiology and would help to prevent oversimplified interpretations of results by the media and the public.
The relationship between bladder cancer, occupation and exposure to a number of occupational agents was assessed in a case-control study conducted in the greater Milan area, Northern Italy. The case series consisted of 263 cases (219 males, 44 females) with histologically confirmed invasive bladder cancer, admitted to a network including major teaching and general hospitals in the area under surveillance. The controls were 287 patients (210 males, 77 females), admitted for acute, non-neoplastic or urinary tract diseases to the same network of hospitals. Cases more frequently reported occupation in dyestuff production (relative risk (RR) = 4.6), painting/spraying work (RR = 1.8), chemical industry (RR = 1.7), pharmaceuticals (RR = 1.7) and coal/gas production (RR = 3.1). Only for dyestuff production however, was the excess statistically significant. There was no association with agriculture or related activities, rubber manufacturing, printing, the petroleum industry, food processing and mechanics. In relation to exposure to occupational agents, significant positive trends in risk were observed for dyes/paints (RR = 4.8 for greater than ten years of exposure), herbicides (RR = 4.1), chemicals (RR = 2.4) and gases/fumes (RR = 4.8). No association was found with metals or metal dusts, plastic resins or glues, oil, wood dust, solvents or benzene, asbestos, electricity or radar and coal tar. Besides confirming the well known association between bladder cancer risk and dyestuff production and, to a lesser extent, a wide spectrum of chemical-related activities, this study provides statistically significant evidence of an independent role of herbicides on the risk of bladder cancer.(ABSTRACT TRUNCATED AT 250 WORDS)
Family history of colorectal cancer interacts with environmental risk factors of colon cancer.
It has long been suggested that subjects diagnosed with cutaneous malignant melanoma (CMM) have an excess rate of subsequent neoplasms. To provide further quantitative information on the issue, we have considered 1,780 histologically confirmed CMM diagnosed between 1974 and 1994 by the Cancer Registries of the French-speaking Swiss Cantons of Vaud and Neuchâtel (760,000 inhabitants) and followed up to the end of 1994 for the occurrence of a second primary. A total of 194 neoplasms was observed vs. 111.7 expected, corresponding to a standardized incidence ratio (SIR) of 1.7 [95% confidence interval (CI) 1.5–2.0]. When skin cancers were excluded, 87 subsequent neoplasms were observed vs. 84.9 expected (SIR 1.0). Significant excess rates were observed for basal cell (SIR 4.4), squamous cell (SIR 3.1) and melanoma (SIR 4.7) of the skin, as well as for prostatic cancer (SIR 2.1). The increased rates of subsequent skin cancer were somewhat larger in males, whereas all the SIRs were systematically greater below age 60. The SIRs of subsequent skin cancer remained above unity for 5 years or longer since diagnosis of CMM, in the absence of a clear pattern in trend with time since diagnosis. The cumulative incidence following CMM was 3% for CMM, 4% for squamous cell and 14% for basal cell carcinoma 20 years after diagnosis of CMM. Our results confirm that patients diagnosed with CMM have excess risks of subsequent melanoma and non-melanomatous skin neoplasms which justify focused prevention and surveillance of skin lesions in these patients. Subjects with CMM do not have any appreciable overall excess of non-skin neoplasms, even after long-term follow-up. Int. J. Cancer 72:776–779, 1997. © 1997 Wiley-Liss, Inc.
A recent article by Anderson et al.1 published in Cancer reported the results from a cohort study including 472 women with a history of Stein-Leventhal syndrome (SLS). No association emerged between SLS and risk of breast carcinoma (14 cases overall). SLS is characterized by menstrual irregularities, infertility, a high level of androgens, and obesity, which are clinical and hormonal conditions associated with the risk of breast carcinoma. Other studies have shown inconsistent data regarding the association between SLS and the risk of breast carcinoma.2, 3 Thus, it is also important to examine further the issue in consideration of the recent hypothesis of an association between SLS and insulin resistance and hyperinsulinemia4 (i.e., possible risk factors for breast carcinoma). Therefore, we analyzed data from a large case-control study on breast carcinoma conducted between June 1991 and February 1994 in six Italian areas: the provinces of Pordenone and Gorizia, the urban areas of Milan and Genoa, the province of Forlì in northern Italy, the province of Latina near Rome in central Italy, and the urban area of Naples in southern Italy. Study methods have been described previously.5, 6 The cases studied were 2569 women (median age 55 years; range, 23-74 years) with incident (i.e., diagnosed within 1 year before the interview), histologically confirmed breast carcinoma who were admitted to the major teaching and general hospitals in the areas under surveillance. The controls were 2588 women (median age, 56 years; range, 20-74 years) who were admitted for acute conditions (22% for traumas, 33% for other orthopedic disorders, 16% for acute surgical conditions, 18% for eye disorders, and 12% for other diseases). The structured questionnaire included information on personal characteristics and habits, education and other socioeconomic factors, general lifestyle, habits such as smoking, alcohol, and coffee consumption, a food frequency section, and menstrual and reproductive history. A past history of 23 selected medical conditions or procedures also was elicited7; among these, information was collected on history of SLS Table 1. In this data set women with breast carcinoma were significantly more educated, reported fewer full term pregnancies, and were more often premenopausal than control subjects.4 Body mass index was associated inversely with breast carcinoma risk in premenopausal women, but was directly associated in postmenopausal women.6 To allow for the possible confounding or modifying effect of these factors in the computation of the odds ratio (OR) of breast carcinoma in women with SLS, we have included in the regression equations, in addition to design variables (i.e., study area and age in quinquennia), years of education (< 7, 7-11, and ≥ 12 years), parity (0, 1, 2, 3, and ≥ 4 births), quintile of body mass index, and, when appropriate, menopausal status. The distribution of cases and controls according to history of SLS and menopausal status is shown in Table 1. A total of 14 cases (0.5%) and 16 controls (0.6%) reported a history of SLS. The OR of breast carcinoma in women with a history of SLS was 0.8 (95% confidence interval [CI] 0.4-1.7) when the whole series was considered, and 1.0 (95% CI, 0.4-2.8) and 0.8 (95% CI, 0.3-2.4), respectively, when the analysis was conducted in strata of premenopausal and postmenopausal status. The main limitation of this data set is that the diagnosis of SLS was self-reported. However, interviewers were instructed to check information in clinical records and whenever inconsistencies emerged, to reinterview the cases and controls. Despite this limitation, these data confirm the lack of an association between SLS and the risk of breast carcinoma. A more consistent association between SLS (as a cause of high unopposed estrogen levels before menopause) and risk of endometrial carcinoma has been reported.8 Therefore, hormonal factors may act on breast carcinogenesis in a different and more complex manner. Fabio Parazzini M.D.*, Carlo La Vecchia M.D. , Silvia Franceschi M.D. , Renato Talamini Sc.D.?, Eva Negri M.D.?, Pier Giorgio Crosignani Sc.D.**
During the dramatic circumstances that the region faced in March 2020, pre-existing socioeconomic inequalities substantially widened. With the reorganization of the health system and the availability of vaccines, these disparities returned to the levels recorded before the pandemic.
