Search

Incidence rates and standardized incidence ratios (SIRs) for invasive and in situ cervical cancers diagnosed in 1982-87 were analyzed according to total population size of 148 towns, using data from the population-based Connecticut Tumor Registry. Previous studies have not considered socioeconomic or sociodemographic variables in analyzing cervical cancer rates in urban-rural or population density groups. In multiple regression analyses, increases in SIRs for invasive and in situ cancers from medium-sized to large towns were explained by differences in poverty prevalence or other sociodemographic variables. Nevertheless, cancer screening programs (aimed at preventing invasive cancer through early detection) could be targeted to some extent on the basis of town size category, because larger towns include large numbers of poorer and minority women with higher rates of invasive cervical cancer. SIRs for in situ cancer were significantly elevated in the largest towns (100,000 + total population), and primary prevention programs could target these larger towns with a higher prevalence of poverty. [ABSTRACT FROM AUTHOR]

Two studies have reported that young women with breast cancer face increased risk of early mortality if their first child was male rather than female. An immunological mechanism has been suggested. We sought to confirm these results in a larger, historical cohort study of 223 parous women who were aged <45 years at breast cancer diagnosis during 1983-1987. Subjects were identified through the Maine Cancer Registry. Follow-up data were obtained from hospitals, physicians, and death certificates. Reproductive history data were obtained from the next of kin of the deceased women, birth certificates, physicians, hospitals, and lastly, subjects. With a 7-year follow-up, multivariate modeling found a lower mortality risk in women with a male first child (hazard ratio [HR] 0.51, 95% confidence interval [CI] 0.32-0.81, log-rank comparison). The survival advantage remained for at least 13 years in women with a male firstborn. Thus, previous studies were not confirmed. Mortality risk in young women with breast cancer is not increased by having borne a male first child rather than a female first child.

Twenty years ago the fetal antigen hypothesis was proposed as a potential mechanism by which women are naturally immunized against cancer antigens by antigens from their fetus. Evidence from recent clinical studies shows that a high percentage of parous woman, but not nulliparous women, show evidence of immunization to antigens found on breast, ovarian and endometrial cancer cells. I suggest that this maternal immunization also affects the fetus, causing early immune rejection of fertilized ova that express cancer-related genotypes. Additional cancers, and perhaps even other types of genetic diseases, may be involved in this mechanism.

The authors assessed body mass index (BMI), measured as Quetelet's index (weight in kilograms divided by the square of height in meters), in relation to lung cancer risk in never and former smokers by using data from a population-based, individually matched, case-control study conducted in New York State from 1982 to 1985. To be included in the study, subjects must never have smoked more than 100 cigarettes in their lifetime (never smokers) or not have smoked more than 100 cigarettes during the last 10 years (former smokers). Data on height and weight were complete for 412 of 439 case-control pairs. A positive relation was found between BMI and lung cancer risk for both never smokers (188 case-control pairs) and former smokers (224 pairs). When subjects were combined, those in the eighth (highest) octile (BMI > 30.84) had more than twice the odds of being cases compared with those in the lowest octile (BMI < or =21.26, 95 percent confidence interval: 1.2, 4.4). These study results are consistent with those from studies of BMI and other cancer sites but differ from lung cancer results usually found in predominantly smoking populations.

The authors conducted a population-based case-control study of lung cancer in nonsmoking men and women in New York State from 1982 to 1984. Nonsmokers included both never smokers (45%) and former smokers who had quit at least 10 years before diagnosis/interview (55%). In-person interviews were completed for 437 lung cancer cases and 437 matched population controls. Cases and controls were asked to report on their family history of cancer, as well as smoking status of family members. Cases were significantly more likely than controls to report having a paternal history of any cancer [odds ratio (OR), 1.67] and aerodigestive tract cancers (OR, 2.78); a maternal history of breast cancer (OR, 2.00); a history of any cancer in brothers (OR, 1.58) and sisters (OR, 1.66); and a nearly significant excess of lung cancer (OR, 4.14; P = 0.07), aerodigestive tract cancer (OR, 3.50; P = 0.06), and breast cancer (OR, 2.07; P = 0.053) in sisters. The excess risk in relatives of cases as compared to relatives of controls also was evident in a cohort analysis of the relatives. These results support the hypothesis of a genetic susceptibility to various cancers in families with lung cancer in nonsmokers.

From 1982 to 1984, the authors conducted a population-based case-control study of lung cancer in men and women nonsmokers in New York State. In-person interviews were completed for 437 lung cancer cases (197 never smokers, 240 former smokers) and 437 matched population controls. Cases and controls were asked to report any history of physician-diagnosed nonmalignant lung disease; cases were more likely than controls to report such a history. Statistically significant associations were found for emphysema (odds ratio (OR) = 1.94, 95% confidence interval (CI) 1.10-3.43), chronic bronchitis (OR = 1.73, 95% CI 1.10-2.72), and the combined endpoint of emphysema, chronic bronchitis, or asthma (OR = 1.82, 95% CI 1.26-2.63). After adjustment for active and passive tobacco smoke exposure, emphysema, chronic bronchitis, and asthma (each condition and the combined endpoint) were significantly associated with lung cancer risk. The risk was more marked for squamous cell carcinomas and for subjects who were diagnosed at older ages, and it remained significant when surrogate interviews were excluded. These results are consistent with the hypothesis that certain prior lung conditions increase the risk of lung cancer in men and women nonsmokers.

Background: In the 1980s, prostate specific antigen (PSA) came into wide use as a prostate carcinoma screening and detection method in the United States. Following the introduction of PSA, the age-adjusted incidence of prostate carcinoma reported by the Surveillance, Epidemiology, and End Results (SEER) program in the United States rose rapidly (from 84.4/100,000 in 1984 to 163/100,000 in 1991). When an increase in incidence is observed following the introduction of a screening method, a subsequent decrease in incidence may be expected as prevalent cases are removed from the population (a cull effect). Incidence rates may also fall due to factors such as decreased intensity of screening. The Utah Cancer Registry data were examined for a decrease in prostate cancer incidence., Methods: We tracked age-adjusted prostate carcinoma incidence trends from the population-based Utah Cancer Registry and compared them with rates from the SEER national registry., Results: A rapid and highly correlated rise in prostate carcinoma incidence has been observed in both SEER and Utah incidence rates between 1988 and 1991, the last year for which SEER data are available. In 1992, Utah incidence rates peaked at 236.2 per 100,000. In 1993 and 1994, Utah incidence rates fell to 195.0, and an estimated 164.0 per 100,000, respectively., Conclusions: Population-based data from the Utah Cancer Registry indicates that the incidence of prostate carcinoma is decreasing rapidly after a similarly rapid increase.

Objectives: Each case of a continuous series of invasive cervical cancer cases was studied with a structured review procedure conducted by an expert panel to assess the reason that it was not detected before it became invasive., Methods: All cases of invasive cervical cancer diagnosed in a 5-year period among Connecticut residents were identified; a screening history and screening outcome were obtained for 72% (481 of 664)., Results: Two hundred fifty women (51.9%) had suboptimal screening. One hundred thirty-seven women (28.5%) had never had a screening test, and their mean age was greater than that of the rest of the study population (64.5 years vs 46.5 years). Of the 344 women who had ever had a Pap test, 113 (32.8%) had their last Pap test 5 or more years before their diagnosis of invasive cancer; 52 (15.1%) were not followed up properly; 33 (9.6%) had their last smear misread as normal; and 118 (34.3%) developed cervical cancer within 3 years of their last Pap test., Conclusions: Physicians, nurses, and other care providers need to ensure that woman have timely and accurate screening with proper follow-up, make increased efforts to reach older women, and improve quality control of Pap smear readings.

Several reproductive factors have been found to be associated with breast cancer. Based on an immunologic rather than strictly hormonal interpretation of these observations, we hypothesized that, beyond the observed associations relating to pregnancy per se, multiple marriages would be found to protect women against breast cancer. We obtained cases and controls from linked records from the Utah Cancer Registry and genealogic records. A total of 2,414 women with newly diagnosed breast cancer and 9,138 individually matched controls were included. The unadjusted odds ratio for each marriage after the first was 0.81 [95% confidence interval (CI) = 0.68-0.97]. After adjustment for major reproductive factors, multiple marriages were still found to reduce the risk of breast cancer (odds ratio = 0.86; 95% CI = 0.72-1.03). The findings are consistent with the operation of an immunologic mechanism in the etiology of breast cancer and suggest that the gene/antigen responsible for this effect is quite common in males. Further empirical evaluation of the fetal antigen hypothesis seems warranted.

Background: Several studies have found that daughters born to older mothers have an elevated risk of breast cancer, and an endocrine hypothesis, among others, has been developed to explain these findings. Three recent studies have failed to find a consistent maternal age effect, indicating a need for further exploration of this issue., Purpose: We used Utah breast cancer records linked to genealogical records to investigate maternal and paternal age and other maternal reproductive factors in relationship to the daughter's risk of breast cancer., Methods: The study group consisted of 2414 breast cancer case patients and 9138 individually matched control subjects. Breast cancer diagnoses were ascertained through the National Cancer Institute's Surveillance, Epidemiology, and End Results Program. The case patients and control subjects were born between 1875 and the end of 1947, and the mean age at diagnosis of the case patients was 65.9 years., Results: No consistent effect for maternal or paternal age was found, except possibly among women who were firstborn children (odds ratio [OR] = 1.42 for a 10-year differential in maternal age; 95% confidence interval [CI] = 1.00-2.00). Further examination of the data indicated that mothers of case patients experienced long intervals between marriage and their first birth but not between subsequent births, and they went on to have fewer children. For each year of delay between the mother's marriage and first birth, the odds of breast cancer in the daughter increased 1.05-fold (95% CI = 1.01-1.10)., Conclusions: We found no evidence of a consistent maternal age effect with regard to breast cancer risk in the daughter, but we did find evidence that the mothers of women who go on to get breast cancer have a reproductive pattern that could suggest some form of underlying infertility., Implications: These findings widen the epidemiologic support for the fetal antigen hypothesis, which is an immunogenetic explanation for the relationships between reproductive factors and breast cancer risk. That hypothesis provides strategies for the identification of breast cancer genes and the eventual development of a breast cancer vaccine.

The fetal antigen hypothesis was proposed over a decade ago to explain the small, short-term promotional effect, as well as part of the long-term protective effect that pregnancy has on breast cancer risk. The hypothesis is based on immune interaction between mother and fetus, and postulates that breast cancer genes which a women's fetus inherits from her mate may be indirectly involved in protecting the woman against breast cancer. This article addresses new epidemiological data pertaining to the hypothesis plus the growing evidence that prenatal reproductive factors are important determinants of breast cancer risk. It also suggests that iso-immunization from pregnancy, similar to Rh iso-immunization, can produce immunity against breast cancer. The proposed mechanism provides a plan for development of a vaccine against breast cancer. If the fetal antigen hypothesis is valid for breast cancer, the mechanism behind it is likely to apply to other cancers, even cancers that are found in males, since males can be affected be the prenatal component of this mechanism.

Background: Approximately 15% of all lung cancer deaths in the United States (about 22,350 deaths annually) may not be directly attributable to active cigarette smoking. Consumption of beta carotene, which is derived almost exclusively from intake of fruits and vegetables, has been associated with a reduced risk of lung cancer in smokers. However, studies examining this association in nonsmokers, particularly nonsmoking men, are limited., Purpose: The purpose of this study was to examine whether dietary factors including beta carotene and retinol are associated with a reduced risk for lung cancer in nonsmoking men and women., Methods: A population-based, matched case-control study of lung cancer in nonsmokers was conducted in New York State from 1982 to 1985. Dietary interviews were completed for 413 individually matched case-control pairs of subjects. To determine whether the relationship between dietary intake from specific food groups and lung cancer differed by type of interview, smoking history, sex, age, or histologic type, we examined data on the case-control pairs from each subgroup separately. The intake of beta carotene and retinol was calculated as the weighted sum of the monthly frequencies of consumption of food items containing these nutrients, where the weights correspond to the nutrient content of a typical portion of the food items., Results: Consumption of greens (P for trend < .01), fresh fruits (P for trend < .01), and cheese (P for trend < .05) was associated with a significant dose-dependent reduction in risk for lung cancer, whereas consumption of whole milk (P for trend < .01) was associated with a significant dose-dependent increase in risk. Use of vitamin E supplements was also protective (odds ratio = 0.55; 95% confidence interval [CI] = 0.35-0.85). Increased consumption of the following food groups was associated with a reduction in risk among females: vegetables (P for trend < .025), raw fruits and vegetables (P for trend < .005), and dairy products (P for trend < .025). In males, increased consumption of raw fruits and vegetables was associated with a reduced risk for lung cancer (P for trend < .005). Dietary beta carotene (OR = 0.70; 95% CI = 0.50-0.99), but not retinol (OR = 0.98; 95% CI = 0.82-1.17), was significantly associated with risk reduction., Conclusions: This is the largest study to date of dietary factors and lung cancer in nonsmokers; results suggest that dietary beta carotene, raw fruits and vegetables, and vitamin E supplements reduce the risk of lung cancer in nonsmoking men and women.

We examined the relationship between the survival of women with breast cancer and the gender of their first children using a genealogy-based survival analysis. The study group consisted of 2,155 parous women diagnosed in Utah (United States) with first primary breast cancers (excluding in situ tumors). We calculated hazard rate ratios (HRR) which were adjusted for stage, median survival times, and proportions surviving for three-, five-, and 10-year intervals stratified by age at diagnosis. Median survival among women diagnosed under the age of 45 was 171 months if the first child was female, but only 66 months if the first child was male (HRR = 1.66, 95 percent confidence interval = 1.07-2.57, for male children). For women diagnosed at age 45 or older, all survival times were similar, although women whose first child was male had slightly longer median survival time. These findings suggest that the gender of the first child has a strong influence on survival among women diagnosed under 45 years of age, but not among those diagnosed later in life. Gender of the first offspring may be a useful clinical indicator of prognosis and survival and may provide insights into etiologic and promotional factors for breast cancer.

This article considers maternal use of alcohol during pregnancy, from an epidemiologic perspective. In general, maternal use of alcohol during pregnancy has been associated with a number of effects in offspring including fetal alcohol syndrome, a reduction in birth weight, effects on behavior, and other late effects. It is apparent from the existing literature that the epidemiologic dimensions of these effects are poorly defined and hampered by methodologic problems. Assessment of both outcome and exposure can be difficult, and there is great potential for uncontrolled confounding. Additionally, it is unclear whether the apparent teratogenicity of alcohol results from direct, acute effects of alcohol on the developing fetus, or results from chronic effects of excessive alcohol intake in the mother. This article concludes with recommendations for further research, which may help clarify the complexity of effects associated with maternal use of alcohol during pregnancy.

A retrospective cohort study was conducted to determine whether producing an offspring with a neural tube defect (NTD) is associated with the development of unusual patterns of cancer among the parents in subsequent years. Mothers and fathers who had a child with an NTD in Upstate New York from 1945 to 1955 were followed through 1979 and compared to a control group of parents with normal offspring. Overall mortality and site-specific cancer incidence were not significantly different for case parents versus control parents, with the exception of gastric cancer, which occurred significantly more frequently in case parents. The finding of a familial association between NTDs and gastric cancer, coupled with a parallel decline in rates of these two diseases, suggests a common aetiology, perhaps related to dietary factors.

Data from a large case-control study of breast cancer were examined to test the hypothesis that maternal age at the birth of female offspring is related to the incidence of breast cancer in daughters. Participants were between the ages of 20 and 54 at the time of the study. Based on results for 2,492 parous women who were newly diagnosed with breast cancer and 2,687 parous controls from the general population, a 15-year increase in maternal age was found to be associated with a 29% increase in the risk of breast cancer in daughters. Adjustment for the daughter's age, her own reproductive history, and other potential confounding factors yielded an estimate of 25% for this increase in risk (95% CI, 8% to 46%). The corresponding increase among 499 nulliparous cases and 457 nulliparous controls was 7%, which was not statistically significantly different in magnitude from the increase among parous women. These findings provide evidence for perinatal influences on the subsequent incidence of breast cancer during adulthood. Although specific mechanisms cannot be inferred directly, the results are consistent with the hypothesis that mutations in the genes of the human egg or sperm play a role in the etiology of breast cancer in female offspring.

A 6-month pilot study was conducted to investigate the feasibility of detecting early pregnancy and early fetal loss in a population of healthy women. To ascertain early pregnancy, monthly 3-day morning urine specimens were collected during each woman's late luteal phase. A specific human chorionic gonadotropin (hCG)-radioimmunoassay, using antibodies against the unique carboxy-terminal portion of beta-hCG, was used to determine pregnancy as early as 9 days fetal developmental age. Eight pregnancies have been reported on a clinical basis, one of which ended in a spontaneous abortion. This pilot study demonstrated the feasibility of detecting early pregnancy and early fetal loss, and an epidemiological study is now being planned to estimate the rate of fetal loss in a large population to establish a baseline for early pregnancy loss.

Birth weight was analyzed among singleton live births (N = 665) in Upstate New York in 1974 to women who used oral contraceptives (OC) in comparison to live births to women who used no contraceptives (N = 716), within 11 months prior to last menstrual period (LMP). In addition, birth weight was examined among live births to women who received hormone support therapy (N = 97) and hormone pregnancy tests (N = 75) during pregnancy. There was no evidence for a reduction in mean birth weight, or an increase in frequency of lower weights, among births to OC users, including those who stopped using OC within 2 months of LMP. Generally similar findings held within three maternal age groups (less than 25, 25-29, and 30-39 years). There was no evidence for a reduction in birth weight among offspring of women who received hormone pregnancy tests. Mean birth weight was relatively low among male and female births to women who received hormone therapy for "threatened abortion," but this may reflect the selection of women for such treatment rather than an effect of exogenous hormones on fetal growth.

The uses of available record systems in epidemiologic studies of reproductive toxicology are described with reference to New York State. The available record systems (and relevant reproductive end points) described include: a newborn screening program for metabolic diseases and hemoglobinopathies (relevant to point mutations); chromosome registries and prenatal cytogenetics (for chromosome anomalies); live birth certificates (for birth defects, birthweight, sex ratio, etc); fetal death certificates (for spontaneous fetal deaths); and a statewide cancer registry (for childhood cancers and transplacental carcinogenesis). The uses and limitations of these record systems are discussed, along with examples of their use in descriptive and analytic epidemiologic studies. Descriptive studies outlined include investigations of temporal and geographic trends in birth defects, birth weight, and fetal deaths, with reference to environmental questions (eg, Love Canal, nuclear power plants). Analytic studies described concern parental occupation in relation to specific birth defects (neural tube defects and Down syndrome) and maternal use of contraceptive drugs.

The New York State Department of Health was asked in September 1978 of investigate a cluster of leukemias and lymphomas in a rural town in western New York State of less than 1,000 people. Four cases of these diseases had been diagnosed in the town's population in the previous 10 months. Residents were concerned about environmental hazards such as background radiation and contamination of their water supply. A total environmental study of the area was not feasible or warranted, but certain environmental studies of the area were conducted. No environmental health hazards were identified. Incidence rates for towns in the four-county area (population 281,000) surrounding the study town were analyzed, based on data from the New York State Cancer Registry. These four counties had been severely affected by the flood following the 1972 Hurricane Agnes. Examination of annual leukemia and lymphoma incidence rates for these counties for 1966--77 revealed that the rates for towns in the river valley (population 102,000), but not for nonriver-valley towns, were 20 to 50 percent above the statewide rates for 1972--77. All other cancer rates remained level throughout both periods. An analysis of spontaneous abortion rates for the four counties for 1968--77 showed a significant peak in 1973, but not for the rest of upstate New York. The peak was concentrated in the towns in the river valley. The apparent time-space cluster of leukemias and lymphomas in conjunction with a marked increase in the spontaneous abortion rate suggests an unidentified flood-related environmental exposure.

This study was designed to assess the health risks associated with exposure to radioactively contaminated gold rings. A group of 135 exposed individuals, who were identified through a statewide jewelry screening program, were studied to determine the frequency of carcinoma and other skin problems on the ring finger. Severity of skin problems increased with increasing length of wear. Forty-one of the exposures were associated with mild to severe skin problems. Nine of the individuals studied were diagnosed as having histologically confirmed squamous cell carcinomas at the site of exposure. The incidence of skin cancer on the ring finger was eleven times that expected for men and forty-five times that expected for women. These data indicate that physicians who have patients with skin lesions of the ring finger should be aware of the possibility of exposure to a radioactive gold ring.

This report compares the history of oral contraceptive (OC) use for 715 New York state women who gave birth to malformed infants with the history for a group of 715 matched controls who gave birth to normal children. The case mothers were found to have used the OC after last menstrual period (LMP), or just before conception, slightly more often than the controls. Confidence intervals for the relative risk estimates were wide, and were significantly higher than 1.0 only among the group of women whose children had one or more major structural or anatomical malformations. Cytogenetic abnormalities and hypospadias were not associated with OC use around or after the time of conception. For women who had children with major anatomical or structural malformations, the association with OCs was stronger among the older age group. The authors found no evidence of an increased risk among women who had conveived their child after one or more pill-free menstrual cycles. Several reported studies, including this one, show a predominance of males among malformed offspring whose mothers used OCs during pregnancy. These findings indicate that the association between birth defects and OC use around the time of conception is not large, but the association is not easily reconciled with a non-causal explanation.

A medical surveillance program has been established for 482 persons who were potentially exposed to polychlorinated biphenyls (PCBs), dibenzo-p-dioxins, and dibenzofurans from an electrical transformer fire in a Binghamton, NY office building in 1981. Blood samples were analyzed for serum concentrations of PCBs and for biochemical and hematologic parameters at the time of the fire and 9 to 12 mo later. Firefighters and individuals who were in the building for 25 hr or more were also asked about post-fire symptomatology and examined after 1 yr for disorders of the skin, eyes, liver, and neurologic system. The results indicated that reported exposure was positively related to mean serum PCB levels (p = .004). The means and individual values, however, were within the range reported by other studies of persons with no unusual exposures. Significant correlations were observed between serum PCB concentrations and levels of liver enzymes and lipids, but mean levels of these biochemical parameters were not associated with reported exposure after adjustment for relevant covariables. Approximately one-half of those examined had skin lesions, but no cases of chloracne were detected, and there was no clinical evidence of any other exposure-related systemic disorder. The data suggest that exposure to contaminants from the building did not result in substantial absorption or cause any major short-term health effects.

The possible effects of maternal spermicide use on birth characteristics of offspring were examined in two studies. First, birth weight of offspring was examined in a cohort study of 302 women who reported using spermicides and 716 women who used no contraceptive methods in the year prior to pregnancy resulting in a 1974 live birth (without a malformation) in Upstate New York. There was no evidence that spermicide use prior to the last menstrual period (LMP) had an effect on mean birth weight or on the proportion of lower weights. Mean birth weight of female births was significantly lower in post-LMP spermicide users than in pre-LMP-only spermicide users and no-contraceptive users. In multiple linear regression analyses of birth weight among births to spermicide users, including maternal smoking during pregnancy and other variables, time of discontinuation of spermicide use was an important predictor of female (but not male) birth weight. In the second (case-control) study of 715 Upstate New York births with selected birth defects and 715 control births (matched on maternal age and race), no significantly increased relative risks were associated with maternal spermicide use prior to LMP or after LMP. Based on small numbers, relative risks for post-LMP spermicide use were greater than 1.00 for hypospadias (8/2 or 4.00, not significant) and for limb reduction defects (6/3 or 2.00, not significant).

The effect of oral contraceptive (OC) use was examined among parous breast cancer cases diagnosed at less than or equal to 45 years of age, and in two different general population control groups of parous women. The primary analysis compared 278 cases with 520 controls. In the secondary analysis, a subgroup of 190 cases were compared with 190 controls from another reference group. In both analyses, ever use of OC, duration of OC use, and OC use prior to first pregnancy showed no significant association with breast cancer risk. Also in both analyses, in women with prior benign breast disease OC use increased the risk of breast cancer, although this finding was statistically significant only in the larger group used in the primary analysis.

A cohort of 574 black Adults (mean age 47 years with a range of 28-73 years) with sickle cell trait (haemoglobin AS) which had been age and sex matched with 1148 controls (haemoglobin AA) was monitored for 7 years. We found no evidence of excess mortality or differential causes of mortality between cases and controls. There were no case deaths related to kidney disease, splenic infarcts, or pregnancy. Reports in the literature have been inconsistent with regard to the health effects of sickle cell trait. These findings suggest that discrimination against trait carriers in terms of life insurance premiums and some job opportunities are unjustified.

Various maternal characteristics were examined in case-control studies of limb-reduction defects among live births in Upstate New York (i.e., New York State exclusive of New York City). In a study using birth records of 305 cases and 915 controls born in 1970-81, mean birth weight and gestation length were significantly lower in cases than in controls, and certain pregnancy complications (i.e., abnormal vaginal bleeding and diabetes mellitus) were reported more often in cases than in controls. Several subgroups of defects (i.e., isolated defects vs. multiple, long-bone defects vs. other) were also compared with their respective controls. The subgroup with multiple (including nonskeletal) defects had the lowest mean birth weight and length of gestation. In an interview study of 108 cases and 108 controls born in 1968-74, a history of maternal use of thyroid medication was a significant risk factor; possible explanations for this association are discussed.

Using data from 5489 cancer patients and 2647 patients without cancer we investigated whether parental age at the birth of the patient or the patient's rank within his sibship was related to the risk of cancer during adulthood. An increase of 10 years in maternal age was associated with an increase of 24% for the incidence of breast cancer (odds ratio = 1.24%; 95% CI = 1.09-1.41); the corresponding increase for paternal age was 19% (odds ratio = 1.19; 95% CI = 1.07-1.33). There was some evidence that the age of each parent may make an independent contribution to the risk of breast cancer. For certain types of genito-urinary cancers, the risk was higher when the parents were relatively young at the birth of the patient. These cancers included tumors arising in the prostate (odds ratios = 0.71 and 0.55 for maternal and paternal ages, respectively), testis (odds ratios = 0.57 and 0.52), penis (odds ratios = 0.37 and 0.45), kidney (odds ratios = 0.66 and 0.60), and bladder (odds ratio = 0.79 and 0.85). The associations for cancer of the prostate and bladder were stronger among patients who were diagnosed at a relatively young age. No statistically significant effects were found for birth order relative to adult cancers. The authors conclude that environmental factors that affect the parents or that operate in the perinatal period may have stronger influences on the incidence of adult cancers than have been previously recognized.

Nine selected census tracts containing 12 toxic-waste disposal sites with known or suspected lung carcinogens were identified in Niagara County, New York. Analysis of death certificates of 339 lung cancer cases (decedents) and 676 controls who died of other causes excluding respiratory diseases in 1978-1981 showed no association between death from lung cancer and residence in the selected census tracts (odds ratio = 0.95; 95% confidence interval = 0.65-1.38). Analysis of mail questionnaires from surrogate respondents for 209 lung cancer decedents and 417 controls showed no significant association between lung cancer and a history of ever having resided in the selected census tracts (age-adjusted odds ratio = 1.17; 95% CI = 0.78-1.76) and no significant interaction between such residence and cigarette smoking. Duration of residence in the selected census tracts did not differ between cases and controls. The limitations of this low-cost study design, in terms of response rates and potential misclassification of exposure, were discussed along with its value in interim studies of potentially hazardous dumpsites (prior to more intensive case-control or other studies using better exposure data).

We examined some characteristics of first pregnancy in 314 female breast cancer patients whose disease was diagnosed at < 45 years of age and who were identified through a population-based cancer registry in New York State and in 314 control women identified through birth records. Preeclampsia was reported less frequently on the birth records of the first recorded pregnancy in the cases than in those of the controls, with a relative risk of 0.28 (0.05 < p < 0.10, from logistic model). We examined this finding on preeclampsia with particular reference to immunologic mechanisms, but further studies are needed on larger samples. Twinning and maternal Rh factor, also relevant to immunologic factors, were not significantly associated with a higher risk of breast cancer. Among the women whose first pregnancy resulted in a live birth, length of gestation (which is correlated with maternal estriol levels) did not differ between the cases and controls.