Likelihood-ratio testing as a diagnostic method for small-sample regression.
Explore the source record for details and available documents.
Likelihood Functions↗
Biomedical subjects
H A Risch
Publications and source records attributed to H A Risch.
Dietary factors and risk of lung cancer: results from a case-control study, Toronto, 1981-1985.
Associations between dietary factors and risk of lung cancer are reported from a study of 839 cases and 772 population-based controls interviewed in metropolitan Toronto between 1981 and 1985. Increased consumption of vegetables is associated with a decreased relative risk of 0.60 (95% confidence limits = 0.40 to 0.88) for those in the highest compared with the lowest quartile. Cholesterol intake is associated with increased risk, but this is restricted to those in the highest quartile for whom the relative risk is 1.58 (95% confidence limits = 1.05 to 2.38) compared with those in the lowest quartile. The results of this study suggest that dietary factors may affect the risk of lung cancer, but identification of the specific constituents involved will require further research.
Adult↗
Mortality from breast cancer after irradiation during fluoroscopic examinations in patients being treated for tuberculosis.
The increasing use of mammography to screen asymptomatic women makes it important to know the risk of breast cancer associated with exposure to low levels of ionizing radiation. We examined the mortality from breast cancer in a cohort of 31,710 women who had been treated for tuberculosis at Canadian sanatoriums between 1930 and 1952. A substantial proportion (26.4 percent) had received radiation doses to the breast of 10 cGy or more from repeated fluoroscopic examinations during therapeutic pneumothoraxes. Women exposed to greater than or equal to 10 cGy of radiation had a relative risk of death from breast cancer of 1.36, as compared with those exposed to less than 10 cGy (95 percent confidence interval, 1.11 to 1.67; P = 0.001). The data were most consistent with a linear dose-response relation. The risk was greatest among women who had been exposed to radiation when they were between 10 and 14 years of age; they had a relative risk of 4.5 per gray, and an additive risk of 6.1 per 10(4) person-years per gray. With increasing age at first exposure, there was substantially less excess risk, and the radiation effect appeared to peak approximately 25 to 34 years after the first exposure. Our additive model for lifetime risk predicts that exposure to 1 cGy at the age of 40 increases the number of deaths from breast cancer by 42 per million women. We conclude that the risk of breast cancer associated with radiation decreases sharply with increasing age at exposure and that even a small benefit to women of screening mammography would outweigh any possible risk of radiation-induced breast cancer.
Adolescent↗
Risk of bladder cancer by source and type of tobacco exposure: a case-control study.
The association between tobacco use and risk of bladder cancer was investigated in a population-based case-control study conducted in Alberta and south-central Ontario, Canada, between 1979 and 1982. In all, 826 histologically-confirmed cancer cases and 792 randomly selected controls, individually matched to cases for age, sex, and area of residence, were recruited into the study. Compared to those who had never smoked cigarettes, males and females who had ever smoked cigarettes had a statistically highly significant 2-fold increase in risk of bladder cancer; for ex-smokers, the risk was intermediate between that for current smokers and never-smokers. There was a dose-dependent increase in risk of bladder cancer with total lifetime cigarette consumption, of similar magnitude for males and females. In males, risk increased with self-reported degree of inhalation in ex-smokers and in current smokers (statistically significant trend), while in females there was no association in current smokers, and a statistically significant inverse association in ex-smokers. Overall, risks of bladder cancer associated with lifetime consumption of plain and filter cigarettes were similar, and there was little evidence to suggest that switching from plain to filter cigarettes was beneficial. Neither passive smoking nor other forms of tobacco consumption (pipes, cigars, chewing tobacco, or snuff) were associated with altered risk of bladder cancer. The population attributable risk for cigarette smoking was about 47% in males and about 33% in females.
Adult↗
An exploratory case-control study of brain tumors in children.
An exploratory case-control study of childhood brain tumors was conducted in southern Ontario between 1977 and 1983, on 74 cases and 138 age- and sex-matched population controls. A significantly elevated risk (perhaps due to early case symptoms) was seen for skull X-rays at least 5 years prior to diagnosis, and for head or neck injuries which required medical attention. However, no evidence of an increased risk appeared for exposure to sick pets or to pesticides, maternal or paternal history of smoking, and various birth characteristics or antenatal exposure of the child, though these have previously been reported to be associated with childhood brain tumors. With respect to the hypothesis that N-nitroso compounds may be involved in the etiology of childhood brain tumors, most exposures of this type were not associated with risk, though a significant positive association was seen for consumption of beer by the mother during pregnancy, and a significant negative association was seen with consumption of fruit juice by the child. Other findings in the present study include an association with developmental problems relating to height and weight and with certain socioeconomic characteristics of the mother. Further investigation of these results in future studies is warranted.
Adolescent↗
Lymphocyte micronuclei and sister chromatid exchanges among Canadian federal laboratory employees.
Blood lymphocytes from 59 federally employed Canadian laboratory workers, 137 nearby office employees, and 11 workers with a mixed employment history were examined for the presence of sister-chromatid exchanges and micronuclei. These laboratory workers were exposed routinely to a host of organic solvents and had previously voiced complaints regarding the adequacy of the ventilation system in their building. Overall, no relationship between laboratory work and either cytogenetic endpoint was observed. The two tests were only weakly correlated, but in a multivariate analysis, both were associated with recent and past smoking. Within-person variation was measured for the sister-chromatid exchange assay, and it and other possible sources of variation are discussed.
Abortion, Spontaneous↗
An application of capture-recapture methods to the estimation of completeness of cancer registration.
Completeness of cancer registration has not been consistently ascertained across different registries. This report describes how capture-recapture methods have been used to estimate completeness at the Ontario Cancer Registry. The method was applied in two fashions; first, using three data sources in a modeling approach: and second, using two data sources and standard, simple capture-recapture methods. The modeling approach is more flexible, since several variables that influence cancer registration can be considered and can be used to identify reporting patterns of different data sources. In the present analysis, estimates of completeness of the registry as a whole were remarkably similar using either two or three data sources, and site-specific comparisons differed by at most 7%. Because of the advantages of capture-recapture methods-estimation of level of completeness, possible comparability of estimates across different registries, and versatility to consider other determinants of cancer registration-a plea for greater use of these methods in cancer registration is made.
Data Collection↗
Dietary factors and the incidence of cancer of the urinary bladder.
A case-control study of diet and bladder cancer was conducted during 1979-1982 in Edmonton, Calgary, Toronto, and Kingston, Canada. A total of 826 histologically verified cancer cases were individually matched by age, sex, and area of residence to 792 randomly selected population controls. Subjects were interviewed concerning their histories of exposure to a number of dietary factors, including table-top artificial sweeteners, low calorie foods and drinks, beverages containing caffeine or ethanol, and certain other food items. Also, subjects provided information on their past medical, occupational, and residential histories, in addition to their exposures to tobacco and other life-style factors. For the analysis, conditional logistic regression methods were used. Under adjustment for cumulative lifetime amount of cigarette smoking, the dietary factors, with little exception, were not associated with significant alteration of risk for bladder cancer. In particular, ever regular use of artificial sweeteners did not appear to be associated with increased risk, either among men (odds ratio = 0.95, p = 0.70) or among women (odds ratio = 1.15, p = 0.53). However, daily intake of cholesterol, calculated from reported frequencies of consumption of nine relevant food items, suggested a mild increase in risk; the odds ratio estimate of trend was 1.07/100 mg average daily intake (i.e., 1.07(5.5) = 1.45 for 550 mg cholesterol per day, as might be consumed in one egg; p = 0.009). A history of diabetes mellitus of onset after age 20 years also seemed to be associated with increased risk of bladder cancer (odds ratio = 1.65, p = 0.019), but this increase did not appear to be the result of use of insulin or other medications, or use of artificial sweeteners or low calorie foods. Thus, this study tends to confirm reports of a lack of association between use of artificial sweeteners and subsequent risk of bladder cancer.
Adult↗
Risk factors for spontaneous abortion and its recurrence.
Pregnancy histories of women interviewed as normal population controls during 1974-1981 in four case-control studies in the US and Canada were examined to identify risk factors for the occurrence of miscarriage. In total, 2,068 ever-gravid women aged 20-79 years at interview (mean age, 50.3 years) described 6,282 pregnancies, including 805 miscarriages. The roles of previous pregnancy history, age at pregnancy, and other factors were evaluated using relative risk binomial regression methods (similar to logistic regression). Risk of miscarriage during a given pregnancy was found to increase directly with the number of previous miscarriages (the risk was closely approximated by (1 + number of prior miscarriages)1.01), but appeared to be unrelated to the order of miscarriages within all previous pregnancies. Maternal age was also highly related to risk after controlling for gravidity and previous miscarriages, with doubled risk (compared with age 20 years) seen for pregnancies in women older than age 40 years. Risk of miscarriage did not appear to be associated with years since previous pregnancy, height, weight or obesity, use of oral contraceptives within one year before pregnancy, or duration of oral contraceptive use. A slight increase in risk was seen for women who had ever regularly smoked cigarettes (relative risk = 1.14, 95 per cent confidence limits = 1.00, 1.30). Thus, the levels of risk of miscarriage found in this analysis are similar to those of previous studies, and the analytic methods suggest how age, obstetric history, and other factors can be simultaneously examined for associations with such risk.
Abortion, Spontaneous↗
Occupational factors and the incidence of cancer of the bladder in Canada.
During 1979-82, a case-control study of occupational factors and urinary bladder cancer was conducted in Edmonton, Calgary, Toronto, and Kingston, Canada. A total of 826 histologically verified cases of cancer were individually matched by sex, age, and area of residence to 792 randomly selected population controls. Subjects were specifically asked about employment in several industries thought relevant to risk of bladder cancer. Information was also obtained on lifelong occupational history, with special attention given regarding exposures to fumes, dusts, smoke, and chemicals. In addition, subjects provided data on past medical and residential history, on intake of certain dietary items, and on exposure to tobacco and other lifestyle factors. Conditional logistic regression methods were used for the analysis. Under adjustment for cumulative lifetime cigarette consumption, it appeared that for both men and women, most of the occupational factors examined were not associated with significant alteration in risk of bladder cancer. For exposures during the period eight to 28 years before diagnosis, however, raised risk was suggested for men employed at least six months in the chemicals industry (odds ratio = 2.37, p = 0.004), in dye manufacturing or the dyeing of cloth (OR = 3.62 and 4.63, p = 0.041 and 0.035, respectively), as tailors (OR = 3.85, p = 0.015), or in jobs in which contact with diesel or traffic fumes occurred (OR = 1.69, p = 0.0008). Increased risk was also seen for men occupationally exposed to tars or asphalt (OR = 3.11, p = 0.019). This study then, at least for men, supports perhaps a few of the suspect industries as related to risk of bladder cancer.
Adult↗
An exploratory case-control study of brain tumors in adults.
An exploratory study of brain tumors in adults was carried out using 215 cases diagnosed in Southern Ontario between 1979 and 1982, with an individually matched, hospital control series. Significantly elevated risks were observed for reported use of spring water, drinking of wine, and consumption of pickled fish, together with a significant protective effect for the regular consumption of any of several types of fruit. While these factors are consistent with a role for N-nitroso compounds in the etiology of these tumors, for several other factors related to this hypothesis, no association was observed. Occupation in the rubber industry was associated with a significant relative risk of 9.0, though no other occupational associations were seen. Two previously unreported associations were with smoking nonfilter cigarettes with a significant trend and with the use of hair dyes or sprays. The data do not support an association between physical head trauma requiring medical attention and risk of brain tumors and indicate that exposure to ionizing radiation and vinyl chloride monomer does not contribute any appreciable fraction of attributable risk in the population studied. The findings warrant further detailed investigation in future epidemiologic studies.
Adult↗
Consumption of precursors of N-nitroso compounds and human gastric cancer.
It has been hypothesized that dietary nitrate and nitrite are converted in the stomach to nitrous acid, which reacts with secondary amines and amides to form nitrosamines and nitrosamides, compounds frequently demonstrated to be carcinogenic in animals, and that vitamins C and E inhibit N-nitroso product formation by chemically reducing nitrous acid. This hypothesis and others were tested in a case-control study (controls were individually matched by age, sex and area of residence), utilizing a standardized, quantitative, dietary history questionnaire interview. Daily nutrient consumption values were calculated from interview responses, and continuous conditional logistic regression was used for the data analysis. Significant findings are as follows: (1) Average daily consumption of nitrite, chocolate and carbohydrate was associated with increasing trends in risk. (2) While citrus fruit intake appeared to be somewhat protective, any protective effect of vitamin C intake was less apparent, and of vitamin E, not at all apparent. (3) Consumption of dietary fibre was negatively associated with gastric cancer risk. These findings appear to implicate a number of dietary components, including nitrite consumption, in the genesis of gastric cancer in humans.
Adult↗
Changes in breast self-examination behavior in a cohort of 8214 women in the Canadian National Breast Screening Study.
A study cohort of 8214 women was formed from all participants in the Canadian National Breast Screening Study who attended their first three screens after the introduction of a structured evaluation protocol involving eight criteria for breast self-examination (BSE). Self-reported BSE frequency was collected by questionnaire at all three screens. Breast self-examination instruction preceded breast exam at each screen but BSE evaluation occurred only at the second and third screen. Reported monthly BSE frequency increased from 18% to 51% to 55% on first, second, and third screens, respectively. The proportion claiming to do no BSE fell correspondingly: 52%, 16%, and 11%. Compliance with each of the seven other BSE criteria was significantly greater at the third screen than at the second. The mean number of these criteria performed at screen 3 was 5.1 compared with 4.4 at screen 2; older women performed as well as younger. The mean number of criteria performed was associated with BSE frequency at screen 3. The authors conclude that women's BSE behavior can be altered, and that integration of BSE evaluation and instruction into routine medical exams seems feasible and potentially useful.
Adult↗
The effect of cortical ablation on afferent activity in the cat somatosensory system.
The modulatory influence of primary S1 cortex upon afferent activity in the somatosensory system was examined in acute cats. Modulatory influences due both to prior afferent activity and efferent activity on peripherally evoked neural responses in thalamic n. ventralis posterolateralis (VPL) and the dorsal column nuclei (DCN) were analyzed. The right superficial radial nerve was stimulated with suprathreshold pulses applied randomly in time. Evoked activity was recorded from ipsilateral DCN, contralateral VPL, and contralateral S1 cortex before and after ablation of S1 cortex. The data were analyzed using a modified functional power series. The responses were characterized by first- and second-order kernels computed by cross-correlation. S1 ablation reduced or eliminated the second positive peak in the VPL response. Incomplete S1 ablation produced a graded effect on this peak. In addition, S1 ablation eliminated the negative notch on the leading edge of the response observed in the nucleus cuneatus. The results from VPL suggest that the second VPL positive peak is produced by a tonic S1 influence on another region of the central nervous system that is coupled to VPL, rather than by a direct VPL-cortical-VPL reflex loop. The results from the DCN confirm earlier studies suggesting a cuneate-cortical-cuneate reflex loop.
Afferent Pathways↗
A non-linear analysis of afferent modulatory activity in the cat somatosensory system.
The effects of prior input on afferent activity in the somatosensory system of the cat have been studied using both twin-pulse experiments and random stimulus trains for which evoked activity is characterized by a set of mathematical functions known as kernels. The kernels quantitatively characterize transformational processes occurring in the system. The first-order kernel is analogous to a classical averaged evoked potential, but to temporally interacting stimuli; the second-order kernel may be likened to a generalized recovery function. Responses to superficial radial nerve stimulation were obtained at sensory (S1) cortex, thalamic n. ventralis posterolateralis (VPL), and the nucleus cuneatus of the dorsal column nuclei (DCN). Comparison of first-order kernels with averaged evoked potentials showed good agreement in chronic awake, chronic anesthetized, and acute anesthetized animals. Second-order kernels indicated complex interactions with an overall depression of subsequent response amplitudes. Comparison between second-order kernels and the results of the two-pulse experiments at both S1 cortex and VPL showed good agreement for early latency response components, but significant deviations for later components suggest higher than second-order non-linearities for this system.
Animals↗
Dietary factors and the incidence of cancer of the stomach.
A case-control study of diet and stomach cancer was conducted during 1979-1982 in Toronto, Winnipeg, and St. John's Canada. Two hundred forty-six histologically verified cancer cases were individually matched by age, sex, and area of residence to 246 randomly selected population controls. Daily nutrient consumption values were calculated from quantitative diet history questionnaire data through use of the US Department of Agriculture Food Composition Data Bank, which was extended and modified for Canadian items. For the analysis, continuous conditional logistic regression methods were used. It was found that consumption of dietary fiber was associated with decreased risk of gastric cancer; the odds ratio estimate of trend was 0.40/10 g average daily intake of fiber (i.e., 0.40(1.5)/15 g, etc.) (p less than 10(-8)). Also, average daily consumption of nitrite, chocolate, and carbohydrate was associated with increasing trends in risk, with odds ratio estimates, respectively, 2.6/mg (p less than 10(-4)), 1.8/10 g (p less than 10(-4)), and 1.5/100 g (p = 0.015). While citrus fruit intake appeared to be somewhat protective (odds ratio = 0.75/100 g daily average, p = 0.0056), vitamin C intake was less so, and vitamin E not at all. Thus, a number of dietary components seem to be implicated in the pathogenesis of stomach cancer.
Adult↗
Lung cancer in women in the Niagara Region, Ontario: a case-control study.
A case-control study of the etiology of lung cancer in women was conducted in the Niagara Region of Ontario, because of local concerns about a high incidence of lung cancer. 51 female patients with lung cancer and 45 matched controls were interviewed. Information was collected about active and passive smoking, occupation and residential history. There was a strong association between active cigarette smoking and lung cancer (ever/never odds ratio 10.0; p less than .001) and 85% of the cases of lung cancer were attributed to active cigarette smoking. No other factors were significantly associated with lung cancer; there was weak evidence of an association between urban environment during childhood and lung cancer (p = 0.07). Associations between lung cancer and air pollution, and residential history, were not demonstrated, contrary to public perception. Thus, a previously reported excess of lung cancer in Niagara females is most likely attributable to cigarette smoking.
Air Pollution↗