Changing cigarette habits and bladder cancer risk: a case-control study

With the use of data from the 8,764 subjects in the National Bladder Cancer Study, the separate contribution of various aspects of a person's cigarette smoking history to his increased risk of bladder cancer was estimated. These estimates have not been previously available, owing to the smaller sizes of earlier studies. Our data indicated that people who have only smoked unfiltered cigarettes have higher risks than those who have only smoked filtered cigarettes but that people who have switched from unfiltered to filtered have experienced no reduction in risk. Our data also indicated that smoking cessation substantially reduced the risk. The former smoker appeared to benefit both because he stopped adding to the burden of irreversible damage and because he ceased being exposed to some reversible hazard. Thus the former smoker had a lower risk than the current smoker even though they had smoked the same number of cigarettes daily for the same number of years, but the former smoker's risk remained higher than the risk of a person who never smoked. Our data suggest that one-half of the bladder cancer occurring among men in the United States and one-third of that among women is caused by cigarette smoking.

Myoglobin recombination at low temperature. Two phases revealed by Fourier transform infrared spectroscopy

The low-temperature recombination of CO with myoglobin was studied using Fourier transform infrared spectroscopy. The bound state of carbon monoxide myoglobin has two distinct conformers observed at 1926 and 1945 cm-1 with an intensity ratio of 1 to 8. The recombination of these bands after complete photolysis at 10 K followed by a temperature jump shows distinct kinetics for the two bands. Although both bands apparently follow the nonexponential kinetics originally described by Frauenfelder and co-workers (Austin, R., Beeson, K., Eisenstein, L., Frauenfelder, H., and Gunsalus, I. (1975) Biochemistry 14, 5355-5373), the 1926 cm-1 band does not appear appreciably below 70 K. In fact, after 20 min of recombination at 70 K the 1945 cm-1 band is fully recovered, while no detectable amount of the 1926 cm-1 band is present. This is the first association of a spectroscopic marker of protein substructure with reaction kinetics.

Menopausal oestrogens and breast cancer risk: an expanded case-control study

A study among 1960 post-menopausal breast cancer cases and 2258 controls identified through a nation-wide screening program enabled evaluation of effects of oestrogen use on breast cancer risk. Ever use was not associated with increased risk (RR = 1.0), but a significant trend was observed with increasing years of use, with users of 20 or more years being at a 50% excess risk. Elevations associated with long-term use were apparent across all menopause subgroups (natural, ovaries retained, ovaries removed). Hormones exerted particularly adverse effects in those initiating use subsequent to a diagnosis of benign breast disease, particularly long-term users (RR = 3.0, 95% CI 1.6-5.5). There was also some indication that effects predominated among the lower stage tumours, an observation similar to that observed for endometrial cancer. These findings support a role for oestrogens in the aetiology of breast cancer, although risk appears to be enhanced only after extended periods of use, and not to the extent observed for other hormonally-sensitive tumours.

Long-term use of oral contraceptives and risk of invasive cervical cancer

To evaluate the relationship between use of oral contraceptives and risk of invasive cervical cancer, a case-control study involving 479 patients and 789 population controls was undertaken in 5 geographic regions of the US. Initially, the relationship was obscured by confounding variables, particularly the interval since last Pap smear. Control for this variable as well as for sexual and sociodemographic factors revealed an RR of 1.5 overall, with long-term users (5 or more years) being at a 2-fold higher risk than non-users. Pill associations prevailed for both adenocarcinomas and squamous-cell tumors, and risks were highest for those using pills containing high estrogen potencies. In addition, there was some evidence that pill associations were most pronounced among women who had never used barrier methods of contraception or who had histories of genital infections, suggesting that oral contraceptives may act as co-carcinogens with transmissible agents. Our findings provide further evidence that long-term use of oral contraceptives may have a carcinogenic effect on cervical epithelium, but emphasize the need for careful evaluation of confounding influences.

Agricultural herbicide use and risk of lymphoma and soft-tissue sarcoma

A population-based case-control study of soft-tissue sarcoma (STS), Hodgkin's disease (HD), and non-Hodgkin's lymphoma (NHL) in Kansas found farm herbicide use to be associated with NHL (odds ratio [OR], 1.6; 95% confidence interval [CI], 0.9, 2.6). Relative risk of NHL increased significantly with number of days of herbicide exposure per year and latency. Men exposed to herbicides more than 20 days per year had a sixfold increased risk of NHL (OR, 6.0; 95% CI, 1.9, 19.5) relative to nonfarmers. Frequent users who mixed or applied the herbicides themselves had an OR of 8.0 (95% CI, 2.3, 27.9) for NHL. Excesses were associated with use of phenoxyacetic acid herbicides, specifically 2,4-dichlorophenoxyacetic acid. Neither STS nor HD was associated with pesticide exposure. This study confirms the reports from Sweden and several US states that NHL is associated with farm herbicide use, especially phenoxyacetic acids. It does not confirm the case-control studies or the cohort studies of pesticide manufacturers and Vietnam veterans linking herbicides to STS or HD.

Carotenoid intake, vegetables, and the risk of lung cancer among white men in New Jersey

A population-based incident case-control study of lung cancer in white males was conducted during 1980-1981 in six high-risk areas in New Jersey. Interviews were completed for 763 cases and 900 controls. To assess whether dietary intake of carotenoids, preformed retinol, or total vitamin A influences the risk of lung cancer, the authors asked the respondents about the usual frequency of consumption, approximately four years earlier, of 44 food items which provide 83% of the vitamin A in the US diet and about the use of vitamin supplements. The men in the lowest quartile of carotenoid intake had a relative risk of 1.3 compared with those in the highest quartile after adjusting for smoking. No increase in risk was associated with low consumption of retinol or total vitamin A. Intake of vegetables, dark green vegetables, and dark yellow-orange vegetables showed stronger associations than did the carotenoid index; the smoking-adjusted risks of those in the lowest quartiles of consumption of these food groups reached relative risks of 1.4-1.5 compared with the risks of those in the highest quartiles. The protective effect of vegetables was limited to current and recent cigarette smokers; the smoking-adjusted relative risks for low consumers reached 1.7, 1.8, and 2.2 compared with the risks for high consumers for vegetables, dark green vegetables, and dark yellow-orange vegetables, respectively. The reduction in risk with vegetable intake was most apparent for squamous cell carcinomas, but it extended to adenocarcinomas and most other cell types when only current and recent smokers were analyzed. This protection among current and recent smokers is consistent with the model that vegetable intake prevents a late-stage event of carcinogenesis. Consumption of dark yellow-orange vegetables was consistently more predictive of reduced risk than consumption of any other food group or the total carotenoid index, possibly because of the high content of beta-carotene relative to other carotenoids in this particular food group.

Serum retinol and retinol-binding protein levels do not predict subsequent lung cancer

Retinol and retinol-binding protein levels were measured in sera previously obtained, and stored in the frozen state, at multiphasic health checkups from 151 persons subsequently found to have lung cancer (cases) and 302 persons who remained free of cancer (controls). Two controls were matched to each case for sex, skin color, age, date of multiphasic health checkup, and aspects of the smoking habit. Mean levels in cases and controls were, respectively, retinol: 82.17 and 82.37 micrograms/dl (p = 0.93), and retinol-binding protein: 6.04 and 6.00 mg/dl (p = 0.81). Mean differences between cases and controls were, retinol: 0.195 micrograms/dl with 95% confidence limits, -3.91 and 4.30 micrograms/dl; retinol-binding protein: -0.033 mg/dl with 95% confidence limits, -0.31 and 0.24 mg/dl. No significant trend in relative risk of lung cancer was observed when the retinol or retinol-binding protein distribution was divided into quintiles. No significant associations were observed in subgroups based on age, sex, histologic type of cancer, cigarette consumption, or interval between blood drawing and cancer diagnosis. In this large study, retinol and retinol-binding protein levels were not useful in predicting the subsequent development of lung cancer.

Truck driving and bladder cancer mortality in rural New England

The relationship between truck driving and bladder cancer mortality was investigated in a case-control study in New Hampshire and Vermont. In-person interviews were conducted with the next-of-kin of 325 bladder cancer cases and 673 controls who died during 1975-79. There were 35 cases and 53 controls who had ever been employed as truck drivers [odds ratio (OR)=1.5, 95% confidence interval (CI)=0.9, 2.6]. There was a statistically significant, but inconsistent, positive association between number of years of truck driving and the OR's, rising to 2.3 (1.2,4.1) for 5 years or more of truck driving. Risk was greatest in men who began driving in the 1930's and 1940's (OR=2.6, 95% CI=1.3,5.1) and among residents of two of the most urbanized counties (OR=3.0, 95% CI=1.2,7.4). The association of bladder cancer with truck driving was unaffected by control for possible confounding factors, such as cigarette smoking and coffee drinking. It was not possible to determine whether the risks associated with truck driving were specifically due to diesel fumes. Truck drivers reporting diesel exposures had an OR=1.8 (95% CI=0.5,7.0), but those without diesel exposure still had an OR=1.5 (95% CI=0.8,2.7). Twenty-six cases and 39 controls reported exposure to diesel emissions in any occupation (OR=1.5, 95% CI=0.8,2.8), and a significant duration-response relationship was seen, rising to threefold for those employed in such jobs for 30-39 years.

Bladder cancer risk and pipes, cigars, and smokeless tobacco

Interview data from 2982 patients with bladder cancer and 5782 controls selected from the general population were used to assess the effects of non-cigarette tobacco use on bladder cancer risk. Compared to men who had never smoked, those who had smoked pipes but not cigars or cigarettes had a relative risk estimated at 1.23 (95% confidence interval [CI] = 0.75-2.00). Those who smoked cigars but not pipes or cigarettes were estimated to have a relative risk of 1.33 (95% CI = 0.92-1.94). Little evidence of dose response was observed. The excess relative risk to pipe smokers was limited to those who inhaled deeply.

Dietary carotene and vitamin A and risk of lung cancer among white men in New Jersey

A population-based incident case-control study of lung cancer in white males was conducted during 1980-81 in six high-risk areas of New Jersey. Interviews were completed with 763 cases and 900 controls or with their next of kin. In order to assess whether dietary intake of carotene, preformed retinol, or total vitamin A modified the risk of lung cancer, subjects were asked about their usual frequency of consumption, several years earlier, of 44 food items, which provides 83% of the vitamin A in the American diet, and about their use of vitamin supplements. The men in the lowest quartile of carotene intake had 1.3 the risk (P-value for trend = .05) of those in the highest quartile after adjustment was made for smoking duration and intensity and education. No association was seen for retinol (P-value for trend = .11) or total vitamin A (P-value for trend = .30). The inverse association between carotene intake and lung cancer was most compelling for squamous cell carcinoma, with the smoking-and education-adjusted risk of those in the lowest quartile reaching 1.4 (P-value for trend = .03) the risk of those men in the highest quartile. Risk of lung adenocarcinoma was not related to carotene intake. The reduction in risk of squamous cell lung cancer with increasing carotene intake was noted in pipe and cigar smokers and cigarette smokers of different intensities. Among nonsmokers adenocarcinoma predominated. The inverse association between carotene and risk of squamous cell lung cancer was not especially strong or graded in response; but it was consistent and could be noted in each stratum when the subjects were divided by education, age, or mode of interview (direct vs. next of kin). The results of the other 4 case-control and 3 cohort studies that have looked at diet and risk of lung cancer are not consistent, and the question whether dietary carotene or total vitamin A reduces the risk of lung cancer is not yet resolved.

Estrogen replacement therapy and benign breast disease

The relationship between estrogen replacement therapy (ERT) and the risk of benign breast disease (BBD) was examined among 929 postmenopausal cases and 846 postmenopausal controls identified through a large breast cancer screening program. Prior use of ERT among postmenopausal women was associated with an increased risk of both fibrocystic breast disease [odds ratio (OR) = 1.4; 95% confidence interval (Cl) = 1.1-1.8] and fibroadenoma (OR = 1.6; Cl = 0.8-3.5). The risk of all BBD rose with increasing years of use (OR = 1.9 for greater than or equal to 15 yr of total use) and with years since initial use (OR = 1.6 for greater than or equal to 15 yr since first use). Among users of the conjugated estrogen Premarin, increased risks were found at all but the lowest dose. Finally, risks associated with ERT were higher in women with a bilateral oophorectomy than in other postmenopausal women. These results suggested, overall, an increased risk of BBD associated with ERT.

Exogenous estrogen and breast cancer after bilateral oophorectomy

Estrogen use in 119 women in whom breast cancer developed after surgically induced menopause was compared to use among an equal number of controls matched for age, date of bilateral oophorectomy, and duration of follow-up. No increased risk for estrogen use versus no use was evident (relative risk = 0.7). When the authors examined three measures of estrogen dose--number of chart notations of estrogen use, time since first use, and duration between first and last use--only those with greater than or equal to 5 notations had any significantly elevated risk (relative risk = 2.1; confidence limits 1.2-3.6), and there was a significant trend toward increasing risk with more notations (P = 0.03). Use specifically of conjugated estrogens was also associated with an increasing risk with more notation of estrogen use (P = 0.07). However, the other two measures of dose did not confirm this trend. Matched multiple logistic analysis suggested that number of notations of estrogen use conferred increased breast cancer risk (relative risk = 1.7), in dose-response relationships, but this result could have occurred by chance. Because of a lack of consistency, the generally low and statistically nonsignificant relative risks, and the lack of consistent effect modification in high-risk groups, the authors were unable to demonstrate a clear increased risk of breast cancer associated with replacement estrogen use.

Diazepam use and progression of breast cancer

The relationship between diazepam and breast cancer was evaluated using data from a case-control study of breast cancer, in which 1075 cases and 1146 controls who were participants in a breast cancer screening program were interviewed. Diazepam use was negatively associated with extent of disease and lymph node involvement, and this effect seemed greatest for long-term users of diazepam. It is not certain to what extent these data reflect an ascertainment bias, an association with the reasons for which the drug was prescribed, or chance. Whatever the explanation, the findings do not support a previous contention that diazepam promotes or accelerates breast cancer growth.

Relationship of thyroid disease and use of thyroid supplements to breast cancer risk

An interview study of 1362 breast cancer cases and 1250 controls identified through a multi-center screening program allowed evaluation of effects of thyroid disease and supplementation on breast cancer risk. A previous diagnosis of treated thyroid disease was not associated with an excess risk (RR = 1.0), nor were any specific diagnoses, including hypothyroidism, hyperthyroidism, or goiter. Although based on limited numbers, women with untreated hypothyroidism or goiter had a significantly reduced risk of breast cancer (RR = 0.3, 95% CI 0.1-0.7). Thyroid supplementation for non-disease reasons (primarily weight loss and fertility problems) was associated with a slight elevation in breast cancer risk (RR = 1.2, 95% CI 0.9-1.7), but patterns of risk by duration and latency generally failed to provide evidence of causality. Elevated risks were noted among women who received thyroid medications for fertility problems (RR = 4.2) and among those with a family history of breast cancer (RR = 2.6) or a late age at first childbirth (RR = 2.4), possibly indicating an hormonal interaction.

Colorectal cancer in rural Nebraska

A case-control interview study of colorectal cancer was conducted in two rural counties of eastern Nebraska to determine reasons for the elevated colon cancer mortality rates during 1950 to 1969. Comparison of the information provided by 86 colorectal cancer cases and 176 matched controls (or their next of kin) revealed an increased risk among persons of Czech background, with persons of Bohemian and Moravian extraction predominating in this area. The data suggest an interaction between Bohemian ancestry and certain dietary patterns in the pathogenesis of colon cancer in this region. Colon cancer risk was elevated among commercial beer drinkers regardless of their ethnic background, although Bohemians reported heavier consumption. An excess risk was also associated with intestinal polyps, reported more often by Moravians, and with familial occurrence of gastrointestinal and other cancers. Since 1969, the mortality and incidence rates for colon cancer in this area have declined, possibly as a consequence of acculturation of the American-born descendants of Czech immigrants.

Cancer mortality among patients with Hansen's disease

For the evaluation of cancer risks associated with immunodeficiencies experienced by patients with Hansen's disease (leprosy) and for the assessment of possible adverse effects of dapsone therapy, a follow-up study was conducted of 1,678 patients admitted to the National Hansen's Disease Center in Carville, La., between 1939 and 1977. Overall, no substantial cancer mortality was observed (standardized mortality ratio = 1.3), nor was there any excess among patients exhibiting defects in cellular immunity by virtue of lepromatous forms of the disease. Notable was the absence of any significant excess of lymphoma (5 observed vs. 2.3 expected), despite the predominance of this tumor in certain other immunodeficiency states. Several cancer sites (oral, bladder, and kidney) occurred excessively, but reasons for the elevations were obscure. Although dapsone has been implicated as a carcinogen in laboratory animals, the use of sulfones, including dapsone, did not appear to affect significantly the risk of any cancers in this population.

Design and methods in a multi-center case-control interview study

We conducted a case-control study in ten areas of the United States in which a total of 2,982 bladder cancer patients and 5,782 population controls were interviewed. We employed a variety of existing and new techniques to reduce bias and to monitor the quality of data collected. We review here many of the design elements and field methods that can be generally applied in epidemiologic studies, particularly multi-center interview studies, and explain the reasons for our selection of the methods, instruments, and procedures used.