The difficult breast

Generalized relative and absolute risk models, in which various functions of time and age modify the excess relative or absolute risk of radiation-induced cancer, are fitted to the Japanese atomic bomb survivor cancer incidence data set. Among generalized relative risk models, those in which a product of powers of time since exposure and attained age modify the relative risk provide the best fit. There are indications that the Armitage-Doll model (in its formulation as a generalized relative risk model) provides a poor fit to the data, possibly in part because of increasing age-adjusted cancer incidence rates in the Japanese cohort. Generalized absolute risk models, and in particular models in which either powers of time since exposure and attained age, or powers of time since exposure and age at exposure modify the excess absolute risk, provide a superior fit to any of the generalized relative risk models for all solid cancer sites analysed together. When six cancer subtypes are examined separately, only for respiratory cancers does this finding remain true, and for two other sites (female breast cancer and thyroid cancer) the generalized relative risk model yields a better fit than the generalized absolute risk model.     

Comparison of breast cancer incidence in the Massachusetts tuberculosis fluoroscopy cohort and in the Japanese atomic bomb survivors

Breast cancer has occurred in excess among women exposed briefly to atomic bomb radiation and among those exposed repeatedly over many years to medical radiation for tuberculosis (TB). The excess relative risk of breast cancer incidence in the Japanese atomic bomb survivors, however, is significantly higher (two-sided P = 0.04) than that in the Massachusetts TB fluoroscopy patients. The best estimate of the ratio between the excess relative risk coefficients for the Japanese and Massachusetts cohorts is 2.11 (95% CI 1.05, 4.95). However, this higher relative excess risk is attributable to the lower baseline risk of breast cancer among Japanese women compared with the Massachusetts women, and the excess absolute breast cancer risks in the two data sets are statistically indistinguishable (two-sided P = 0.32). The best estimate of the ratio between the excess absolute risk coefficients among Japanese and Massachusetts women is 0.73 (95% CI 0.41, 1.44). After childhood exposures, an early onset of radiation-induced breast cancer was seen among Japanese atomic bomb survivors but not among the Massachusetts women. There are some indications (two-sided P = 0.04) of differences in the patterns of risk over time since exposure between these groups exposed in childhood. However, in general there are no marked differences between the Massachusetts and Japanese data sets in the age and time distribution of risk of radiation-induced breast cancer. These data provide little evidence for a reduction of breast cancer risk after fractionated irradiation.     

The risk of non-melanoma skin cancer incidence in the Japanese atomic bomb survivors

The latest Japanese atomic bomb survivor non-melanoma skin cancer incidence dataset is analysed and indicates substantial curvilinearity in the dose-response curve, consistent with a possible dose threshold of about 1 Sv, or with a dose-response in which the excess relative risk is proportional to the fourth power of dose, with a turning-over in the dose-response at high doses (> 3 Sv). The time distribution of the radiation-induced excess risk is best described by a model in which the relative excess risk is proportional to a product of powers of time since exposure and attained age. The fits of generalized relative risk models with exponential functions of time and age at exposure (and in particular of attained age) to adjust the relative risk are less satisfactory, as also are the fits of other models in which products of powers of time since exposure, age at exposure and attained age adjust the excess absolute risk. Sensitivity analyses indicate the importance of likely adjustments to the Hiroshima neutron doses for the optimal model parameters, particularly if values of the neutron relative biological effectiveness (RBE) of more than 5 are assumed. If adjustments recently proposed are made to the Hiroshima neutron doses, then using the optimal model (in which excess risk is proportional to the fourth power of dose) the best estimate of the neutron RBE is 1.3 (95% CI < 07.1). However, uncertainties in skin dose estimates for the atomic bomb survivors means that the findings with respect to the neutron RBE and the non-linearity in the dose-response curve should be treated with caution.     

Breast cancer after radiotherapy for skin hemangioma in infancy

Between 1920 and 1959, 9675 women were irradiated in infancy for skin hemangioma at Radiumhemmet, Stockholm. They were exposed to low to moderate doses of ionizing radiation. The mean age at first exposure was 6 months and the mean absorbed dose to the breast anlage was 0.39 Gy (range <0.01-35.8 Gy). The breast cancer incidence was analyzed by record linkage with the Swedish Cancer Register for the period 1958-1986. Seventy-five breast cancers were found [standardized incidence ratio = 1.24; 95% confidence interval (CI) 0.98-1.54] after a mean absorbed dose of 1.5 Gy in the breasts with cancer. The analyses showed a significant dose-response relationship with a linear model estimate for the excess relative risk (ERR) of 0.38 at 1 Gy 95% CI 0.09-0.85). This relationship was not modified significantly by age at exposure or by dose to the ovaries. The ERR increased significantly with time after exposure and for > or = 50 years after exposure the ERR at 1 Gy was 2.25 (95% CI 0.59-5.62). The fitted excess absolute risk (EAR) was 22.9 per 10(4) breast-year gray. The breast absorbed dose and time after exposure were important risk determinants for breast cancer excess risk. Forty to 50 years of follow-up was necessary for the excess risk to be expressed. The study confirms previous findings that the breast anlage of female infants is sensitive to ionizing radiation.     

First analysis of mortality and occupational radiation exposure based on the National Dose Registry of Canada

A cohort mortality study of occupational radiation exposure was conducted using the records of the National Dose Registry of Canada. The cohort consisted of 206,620 individuals monitored for radiation exposure between 1951 and 1983 with mortality follow-up through December 31, 1987. A total of 5,426 deaths were identified by computerized record linkage with the Canadian Mortality Data Base. The standardized mortality ratio for all causes of death was 0.61 for both sexes combined. However, trends of increasing mortality with cumulative exposure to whole body radiation were noted for all causes of death in both males and females. In males, cancer mortality appeared to increase with cumulative exposure to radiation, without any clear relation to specific cancers. Unexplained trends of increasing mortality due to cardiovascular diseases (males and females) and accidents (males only) were also noted. The excess relative risk for both sexes, estimated to be 3.0% per 10 mSv (90% confidence interval 1.1-4.8) for all cancers combined, is within the range of risk estimates previously reported in the literature.     

Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies

The thyroid gland of children is especially vulnerable to the carcinogenic action of ionizing radiation. To provide insights into various modifying influences on risk, seven major studies with organ doses to individual subjects were evaluated. Five cohort studies (atomic bomb survivors, children treated for tinea capitis, two studies of children irradiated for enlarged tonsils, and infants irradiated for an enlarged thymus gland) and two case-control studies (patients with cervical cancer and childhood cancer) were studied. The combined studies include almost 120,000 people (approximately 58,000 exposed to a wide range of doses and 61,000 nonexposed subjects), nearly 700 thyroid cancers and 3,000,000 person years of follow-up. For persons exposed to radiation before age 15 years, linearity best described the dose response, even down to 0.10 Gy. At the highest doses (> 10 Gy), associated with cancer therapy, there appeared to be a decrease or leveling of risk. For childhood exposures, the pooled excess relative risk per Gy (ERR/Gy) was 7.7 (95% CI = 2.1, 28.7) and the excess absolute risk per 10(4) PY Gy (EAR/10(4) PY Gy) was 4.4 (95% CI = 1.9, 10.1). The attributable risk percent (AR%) at 1 Gy was 88%. However, these summary estimates were affected strongly by age at exposure even within this limited age range. The ERR was greater (P = 0.07) for females than males, but the findings from the individual studies were not consistent. The EAR was higher among women, reflecting their higher rate of naturally occurring thyroid cancer. The distribution of ERR over time followed neither a simple multiplicative nor an additive pattern in relation to background occurrence. Only two cases were seen within 5 years of exposure. The ERR began to decline about 30 years after exposure but was still elevated at 40 years. Risk also decreased significantly with increasing age at exposure, with little risk apparent after age 20 years. Based on limited data, there was a suggestion that spreading dose over time (from a few days to > 1 year) may lower risk, possibly due to the opportunity for cellular repair mechanisms to operate. The thyroid gland in children has one of the highest risk coefficients of any organ and is the only tissue with convincing evidence for risk about 1.10 Gy.     

Acquired urinary incontinence in bitches: its incidence and relationship to neutering practices

A five-year cohort study was conducted on bitches chosen by a sample of 233 randomly selected practising veterinary surgeons in the UK, to estimate the incidence of acquired urinary incontinence (AUI) in neutered and entire animals, and to investigate possible risk factors associated with neutering practices. Information was collected using questionnaires, and data on 809 bitches, of which 22 developed AUI, were obtained. The estimated incidence rates in neutered and entire animals were 0.0174 and 0.0022 per animal-year, respectively (95 per cent confidence intervals: 0.0110, 0.0275 and 0.0009, 0.0058, respectively). The relative risk, neutered vs entire, was 7.8 (95 per cent confidence interval: 2.6, 31.5). The attributable proportion(exposed) and population attributable proportion were 87.1 per cent and 63.1 per cent (95 per cent confidence intervals: 61.9 per cent, 95.6 per cent, and 28.3 percent, 88.5 per cent, respectively). An increased risk, significant at the conventional 5 per cent level, was not demonstrated in animals neutered before, vs after, first heat (relative risk: 3.9, 95 per cent confidence interval: 0.8, 10.4), although the result was significant at the 10 per cent level. Removal of the cervix was not shown to be a risk factor in neutered dogs.     

A prospective study of reproductive factors and risk of epithelial ovarian cancer

BACKGROUND: Parity and long term use of oral contraceptives have been associated consistently with a decreased risk of ovarian cancer. However, previous reports of the relationship of other reproductive factors (time since first use or last use of oral contraceptives, age at menarche or menopause, age at first birth) with ovarian cancer have been inconsistent. METHODS: The authors studied these relationships in the Nurses' Health Study, a prospective cohort study of 121,700 female registered nurses aged 30-55 years in 1976 when the study began. From 1976 to 1988, 260 cases of confirmed epithelial ovarian cancer occurred among 1.2 million person-years of follow-up. RESULTS: A statistically significant inverse association was observed between parity and ovarian cancer risk (relative risk [RR] = 0.84; 95% confidence interval [CI] = 0.77-0.91 per pregnancy); age at first birth was not associated independently with risk. In age-adjusted analyses, a significant inverse association was noted between long term use of oral contraceptives and ovarian cancer, which was no longer significant after controlling for other ovarian cancer risk factors (RR with > or = 5 years' use: 0.65; 95% CI = 0.40-1.05). After control for duration of use, a weak nonsignificant inverse association was observed with time since first oral contraceptive use and no independent effect of time since last use. Neither age at menarche nor age at menopause was associated significantly with ovarian cancer risk. CONCLUSIONS: In this large prospective study, parity was the only reproductive factor that had a substantial independent association with ovarian cancer. Long term oral contraceptive use also appeared to have an inverse relationship with ovarian cancer, although this association was of borderline significance (P = 0.11) after adjustment for other risk factors.     

Role of heat shock proteins in heart transplant rejection

STUDY OBJECTIVE: To study secular trends of placental abruption (PA), the effects of demographic variables and the use of cesarean section (CS) associated with PA. DESIGN: A population based cohort study. SETTING: The Medical Birth Registry of Norway. PATIENTS: 9,592 cases of PA of a total of 1,446,154 births notified, i.e. all births in Norway 1967-1991. MAIN RESULTS: The PA proportion was 6.6 per 1000 births of a gestational age of 16 weeks or more, ranging from 5.3 in 1971 to 9.1 in 1990. Birth order two had the lowest proportion and it increased by maternal age. The PA proportion decreased by gestational age from 86.4 per 1000 below 28 weeks to 3.4 in term pregnancies. The PA proportion per 10,000 pregnancies at risk increased from 1.3 in the 28th week to 14.1 in the 42nd week. A secular trend of a decreasing but still high relative risk of PA in SGA-births at any gestational age increased from 1967 through 1991. The relative risk of PA of Apgar score <7 after five minutes, adjusted for gestational age, was 7.8. CONCLUSIONS: Inspite of an increasing CS rate, an increasing proportion of PA was noted from 1967 through 1991. The proportion was lowest for birth order two and increased by maternal age. To an increasing extent, PA births were centralized. SGA, prematurity and asphyxia were major problems associated with PA. A tendency towards larger infants and a decreasing relative risk of PA in SGA-births might be attributable to improvements in antenatal care.     

Radiology in the neonatal intensive care unit: dose reduction and image quality

This paper describes a prospective study of the diagnostic radiation doses received in a neonatal intensive care unit (NICU) for a representative radiological technique used at our institution for a number of years and a "low dose" technique similar to that recommended by the Commission of the European Communities (CEC). A 400 speed film-screen combination was used in both techniques. A total of 363 anteroposterior (AP) chest and abdominal films of 77 neonates were accrued. For each radiograph, the entrance skin dose (FSD), energy imparted (EI) and mean whole body dose were determined. For a neonatal AP chest, there was an 18% reduction in the mean ESD per radiograph from 20.0 muGy for the representative technique to 16.4 muGy for the low dose technique (p < 0.0005). The reduction in the mean EI per radiograph values for the two techniques from 7.9 muJ to 7.1 muJ (10%) was statistically significant at the p < 0.017 level, after compensating for the difference in mean field dimensions between the two patient cohorts. The mean whole body dose per radiograph reduction from 4.4 to 3.5 muGy (20%) was statistically significant at the p < 0.0028 level. It was determined that the ESD and EI could be fitted by an exponential function in the equivalent patient diameter, a single parameter indicative of neonate size. Absolute excess childhood cancer mortality risk per film was estimated using risk factors derived for fetal exposures. A "worst case" absolute excess mortality risk per chest radiograph was estimated to be 1.40 x 10(-7) for the conventional technique and was further reduced to 1.11 x 10(-7) for the low dose technique. A blind comparison of patient-matched film pairs for each technique was performed by three radiologists using criteria similar to those specified by the CEC. No statistically significant difference in clinical image quality was found between the two techniques.     

Risk of benzene-induced leukemia: a sensitivity analysis of the pliofilm cohort with additional follow-up and new exposure estimates

This report updates the risk assessment by Crump and Allen (1984) for benzene-induced leukemia that was used by OSHA (1987) to support its reduction of the permissible exposure limit (PEL) to 1 ppm and that also was the basis for EPA's (1985) interim "unit risk" for benzene. The present study derives new risk estimates using data from follow-up through 1987 (whereas the earlier assessment only had follow-up available through 1978), and using new exposure estimates for this cohort developed by Paustenbach et al. (1992) that account for a number of factors that were unknown or not fully evaluated in earlier exposure assessments. There was a significant excess of acute myelocytic or acute monocytic leukemia (AMML, the only forms of acute nonlymphatic leukemia observed) in this cohort, and this end point also exhibited a strong dose-response trend. AMML was the only hematopoietic or lymphatic cancer that was clearly linked to benzene exposure. However, quantitative estimates of risk based on modeling either AMML or all leukemia differed by only 20%. Differences between the two Pliofilm plant locations in the occurrence of AMML were not statistically significant (.12 < or = p < or = .21) after differences in levels of benzene exposure were taken into account. The Paustenbach et al. exposures predicted a quadratic dose response, based on a measure of exposure that weighted intensity of exposure more heavily than duration of exposure. The best-fitting quadratic models predicted an additional lifetime risk of a benzene-related death from 45 yr of exposure to 1 ppm of between 0.020 and 0.036 per thousand. Statistical confidence intervals (90%) on these estimates were barely wide enough to include risk estimates based on linear dose response models. These linear models predicted risks of between 1.6 and 3.1 per thousand.     

Pre-AIDS mortality in the Edinburgh City Hospital HIV cohort

In this paper, we look at the incidence and predictive factors of pre-AIDS mortality among HIV-infected individuals, and injecting drug users (IDUs) in particular, and compare IDUs with non-IDUs. 627 patients (73 per cent IDUs) of the Edinburgh City Hospital HIV Cohort were enrolled pre-AIDS and followed up until September 1994. Analyses were performed using cumulative hazard and cumulative incidence estimators for a competing risks model, the Cox proportional hazards model and the non-parametric hazard estimator of Fusaro et al. (1993). The effects of age and CD4 T-lymphocyte cell count, progressively depleted during HIV progression, were investigated. 60 deaths occurred in AIDS-free patients during follow-up; 25 were drug-related deaths in IDUs. Pre-AIDS mortality was higher among IDUs than non-IDUs (p = 0.07). The cumulative incidences of pre-AIDS death after five years from enrollment were 11 per cent in IDUs and 6 per cent in non-IDUs; the cumulative AIDS incidences were, respectively, 19 per cent and 32 per cent. After eight years, cumulative pre-AIDS death incidence was 15 per cent among IDUs; cumulative AIDS incidence among IDUs was 35 per cent. Both groups had similar risks of medically-related (non-AIDS)-MRNA-death. Age and CD4 count were both individually predictive of MRNA death (relative risks (RRs); 2.1 per decade of life, p < 0.01; and 1.9 for each 100 cells per 100 microliters lost, p < 0.0001), although when used together age was less significant (RR 1.6, p = 0.07). Neither was statistically significant for drug-related mortality, although hazard may be lower in older individuals and may increase with falling CD4 count. The drug-related mortality was 1.1 per cent: 2.3 per cent in the first two years after enrollment, and 0.4 per cent thereafter. We conclude that older HIV-infected individuals are at greater risk of medically-related death before AIDS. This risk increases as CD4 count declines. Drug-related hazard may be greater in younger individuals and may increase as CD4 counts fall, but neither effect was formally significant.     

Prognostic significance of lymph node dissection in gastric cancer

The results for 162 patients who underwent curative gastrectomy for gastric cancer from January 1988 to June 1994 were analysed statistically with special reference to the effect of lymph node dissection. Median survival was 69.3 months and the overall cumulative 5-year survival rate was 50.2 (95 per cent confidence interval (c.i.) 41.6-58.1) per cent. By univariate analysis age, histology, depth of tumour invasion, node involvement, number of metastatic lymph nodes and type of lymphadenectomy were found to be significant factors related to survival time. Multivariate analysis with the Cox model and stratified for tumour node metastasis stage revealed that only the number of metastatic nodes (P = 0.04) and the extent of lymphadenectomy (P = 0.003) affected survival independently. With respect to D1 lymphadenectomy, the relative risk associated with D2 and D4 lymphadenectomy was respectively 0.61 (95 per cent c.i. 0.34-1.10) and 0.26 (95 per cent c.i. 0.12-0.60). The 5-year survival rate was 28 per cent for patients who had a D1 dissection, 63 per cent for those who had D2 and 68 per cent for those who had D4. These results suggest that extended lymphadenectomy (D2) and especially superextended lymphadenectomy (D4) can improve survival in patients with gastric cancer.     

Risk factors for mortality among hospitalized children with persistent diarrhoea in Pakistan

We evaluated factors associated with mortality among a cohort of malnourished children with persistent diarrhoea (PD) admitted for nutritional rehabilitation with a defined rice-lentil (Khitchri) and yoghurt diet. Of 302 children consecutively admitted with PD, 13 (4 per cent) died, mostly (62 per cent) within 72 h of admission. Univariate analysis of risk factors at admission associated with mortality indicated significantly increased risk of death with severe stunting [relative risk (RR) 3.1, 95 per cent confidence interval (CI) 1.1-9.0], hypoalbuminaemia (RR 4.3, 95 per cent CI 1.5-12.3), stool frequency > 12/day (RR 6.0, 96 per cent CI 2.0-17.6), stool volume > 100 g/kg/day (RR 10.7, 95 per cent CI 3.0-37.6) and severe dehydration (RR 7.5, 95 per cent CI 2.6-21.8). Children who died also had comparatively shorter duration of diarrhoea at admission, and were also associated with higher rate of bacteremia at admission (Fisher's exact test P < 0.01). The logistic regression model evaluating multivariate risk of mortality identified weight-for-age z-score and sepsis as significant risk factors. Our data suggest that severe malnutrition and sepsis are associated with significantly increased risk of mortality in children with PD. Stringent screening for infections and recognition of subgroups with severe malnutrition and severe diarrhoea may improve screening and case management strategies for this disorder.     

Estimation of excess risk from case-control data using Aalen's linear regression model

We introduce methods for statistical inference in Aalen's non-parametric linear regression model of disease incidence (Aalen, 1989, Statistics in Medicine 8, 907-925) from nested case-control data. These methods provide the basis for estimation of excess risk as a linear function of dose and absolute risk for a given exposure history. The methods are illustrated by estimating excess and absolute risks associated with radon exposure and smoking from nested case-control samples from the Colorado Plateau uranium miners cohort.     

Study of the bioequivalence of a new isosorbide dinitrate tablet formulation compared with the standard preparation

Army Chemical Corps personnel who served in Vietnam were among those service personnel with the greatest potential for exposure to herbicides. An earlier evaluation of the mortality experience of 894 Army Chemical Corps Vietnam veterans found a statistically significant excess risk of dying from digestive disease, primarily due to cirrhosis of the liver, and from motor vehicle accidents. That study was expanded to include 2,872 Vietnam veterans who served with the Army Chemical Corps and a comparison cohort of 2,737 veterans who never served in Southeast Asia but who did serve in the same occupational category. The results of the analysis comparing the Vietnam cohort to the non-Vietnam cohort support the earlier finding of a significant excess of deaths from digestive diseases (adjusted relative risk (RR) = 3.88, 95% C.I. = 1.12-13.45) primarily due to liver cirrhosis. Non-significant elevated relative risks were observed for all cancers combined, digestive and respiratory systems cancers, skin cancer, lymphopoietic cancers, and respiratory system diseases. Compared to the mortality rates in the general population, the non-Vietnam Army Chemical Corps veterans had a statistically significant deficit in mortality from all causes combined, which is consistent with a 'healthy selection bias' seen among military populations (SMR = 0.79, 95% C.I. = 0.66-0.94). For the Vietnam veterans, patterns of elevated but nonsignificant SMRs persisted for diseases of the digestive and respiratory systems and for selected cancer sites.     

Trends in mortality of childhood-onset insulin-dependent diabetes mellitus in Leicestershire: 1940-1991

The relative risk of death by calendar date of diagnosis was investigated in a population-based incident cohort of 845 (463 males:382 females) IDDM diagnosed in Leicestershire before the age of 17 years between 1940 and 1989. The mortality status of 844 (99.9%) patients was determined as of the 31 December 1991, representing 14,346 person-years of risk. Trends in relative risk of death were investigated using Cox proportional hazards modelling for within cohort comparisons and age/sex and calendar time adjusted standardized mortality ratios (SMR) using generalized linear modelling for external comparisons. Median age at diagnosis was 10 years (range 3 months to 16 years); median duration of diabetes 15 years (range 1-51 years). Forty-four patients had died (5.2%; median age at death 31 years, range 11-51 years). A further four patients died at presentation (within 24 h) from ketoacidosis and are excluded from all analyses. Calendar date of diagnosis was found to be an important predictor of mortality. Adjusting for attained age there was evidence of a decline in relative risk of death with calendar date of diagnosis of 3.4% (95% CI, 0.005-6.9%) per annum, equivalent to a 32% fall per decade (95% CI, 5-51%), or 84% (95% CI, 21-97) from 1940 to 1989. The data are consistent with a large fall in mortality between the 1940s and 1950s representing over 50% of the total reduction in mortality between 1940 and 1991. Neither sex nor age at diagnosis were significant predictors of mortality. Over the study period 1940-89 the SMR (male and female combined) fell from 981 (541-1556) to 238 (60-953) relative to the general population. This population-based study shows that the prognosis for Type 1 (insulin-dependent) diabetes mellitus has improved markedly over the period 1940-1991.     

The "epidemic" of breast cancer in the U.S.--determining the factors

Breast cancer incidence rates in the United States rose by 24% between 1973 and 1991. Mortality during this period, however, remained stable. Both the 5-year relative survival rate and the rates of in situ and stage I breast cancers have been increasing, while the incidence of later-stage cancers has been decreasing. Increased mammography screening may explain the documented jump in breast cancer incidence rates during the mid-1980s. Differences in the distribution of breast cancer risk factors may account, in part, for the temporal trends in breast cancer incidence. In particular, breast cancer risk factors may vary by birth cohort, including age at menarche, age at first birth, physical activity, obesity, diet, alcohol intake, estrogen therapy, and exposure to environmental organochlorines. After decades of epidemiologic research, a preventive approach to breast cancer that focuses on the physiologic effects of the sex steroid hormones, and their potential interactions with family history, is being carefully formulated.