Preconception vitamin D and miscarriage in a prospective cohort study.

STUDY QUESTION: Is preconception vitamin D level associated with the risk of miscarriage? SUMMARY ANSWER: Preconception vitamin D levels are not associated with the risk of miscarriage in a population of women conceiving naturally. WHAT IS KNOWN ALREADY: In humans, low vitamin D has been associated with prolonged menstrual cycles, delayed ovulation and a lower probability of conception. Animal and in vitro data indicate that vitamin D may affect implantation. STUDY DESIGN, SIZE, DURATION: This prospective time-to-pregnancy study included 362 women who were trying to conceive naturally between 2008 and 2015. PARTICIPANTS/MATERIALS, SETTING, METHODS: This study included participants who had been trying to conceive naturally for 3 months or less at enrollment and aged 30-44 years. A preconception blood sample was collected and 25-hydroxyvitamin D [25(OH)D] was measured. Women who conceived (N = 362) were at risk of a miscarriage from the day of a reported positive pregnancy test until either a participant-reported pregnancy loss or 20 weeks post day of last menstrual period, whichever came first. Gestational age was defined by ovulation. Time to miscarriage (days) or censoring was modeled using a multivariate Cox proportional hazards model. Multiple imputation was performed for missing covariates and missing day of ovulation. MAIN RESULTS AND THE ROLE OF CHANCE: The mean age was 33 years (SD: 3.0 years). Mean 25(OH)D was lower among those who reported their race as African-American and those with a higher BMI. After adjustment for age, race, BMI, education, exercise, alcohol and caffeine intake, compared to the referent group (30-<40 ng/ml), the hazard ratio (HR) and 95% CI for those with a low 25(OH)D level (<30 ng/ml) was 1.10 (CI: 0.62, 1.91). Among participants with a higher 25(OH)D level (≥40 ng/ml), the HR was 1.07 (CI: 0.62, 1.84). LIMITATIONS, REASONS FOR CAUTION: This study was limited by a 25(OH)D measurement at only a single time point. A large percentage of women in this study had sufficient vitamin D levels, which may have limited our power to detect an effect of deficiency. Women in this study were older (30-44 years), and predominantly reported their race as White which may limit generalizability. WIDER IMPLICATIONS OF THE FINDINGS: The findings of this study do not suggest an association between preconception vitamin D and miscarriage. Future research should focus on women at greater risk for miscarriage or in populations at risk for vitamin D deficiency or on supplementation. STUDY FUNDING/COMPETING INTEREST(S): This research was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (Z01ES103333). This research was also supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health (NIH) under award numbers R00HD079659 and R01HD067683. The authors have no conflicts of interest. TRIAL REGISTRATION NUMBER: N/A.

A prospective study of type 2 diabetes, metformin use, and risk of breast cancer.

BACKGROUND: Type 2 diabetes (T2D) has been associated with increased breast cancer risk, but commonly prescribed antidiabetic medications such as metformin may reduce risk. Few studies have investigated T2D and medications together in relation to breast cancer. PATIENTS AND METHODS: Data came from 44 541 Sister Study participants aged 35 to 74 years at enrollment (2003-2009) who satisfied eligibility criteria, followed through 15 September 2017. Information on time-varying, self-reported, physician-diagnosed, prevalent and incident T2D, use of antidiabetic medications, and covariates was obtained from baseline and follow-up questionnaires. Incident breast cancers were confirmed with medical records. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated. RESULTS: During follow-up (median, 8.6 years), 2678 breast cancers were diagnosed at least 1 year after enrollment. There were 3227 women (7.2%) with prevalent and 2389 (5.3%) with incident T2D, among whom 61% (n = 3386) were ever treated with metformin. There was no overall association between T2D and breast cancer risk (HR 0.99; 95% CI, 0.87-1.13). However, T2D was associated with increased risk of triple-negative breast cancer (HR 1.40; 95% CI, 0.90-2.16). Compared with not having T2D, T2D with metformin use was not associated with overall breast cancer risk (HR 0.98; 95% CI, 0.83-1.15), but it was associated with decreased risk of estrogen receptor (ER)-positive breast cancer (HR 0.86; 95% CI 0.70-1.05) and increased risk of ER-negative (HR 1.25; 95% CI, 0.84-1.88) and triple-negative breast cancer (HR 1.74; 95% CI, 1.06-2.83). The inverse association with ER-positive cancer was stronger for longer duration (≥10 year) metformin use (HR 0.62; 95% CI, 0.38-1.01; P for trend = 0.09). Results were supported by sensitivity analyses. CONCLUSION: Our findings suggest that associations between T2D and breast cancer may differ by hormone receptor status and that associations between T2D and ER-positive breast cancer may be reduced by long-term metformin use.

Helicobacter pylori seropositivity is associated with antinuclear antibodies in US adults, NHANES 1999-2000.

Infectious diseases, such as Helicobacter pylori, which produce systemic inflammation may be one key factor in the onset of autoimmunity. The association between H. pylori and antinuclear antibodies (ANA), a marker of autoimmunity, has been understudied. Data from the 1999-2000 National Health and Nutrition Examination Survey were used to evaluate the cross-sectional association between H. pylori seroprevalence and ANA positivity in US adults aged ≥20 years. ANA was measured in a 1:80 dilution of sera by indirect immunofluorescence using HEp-2 cells (positive ⩾3). H. pylori immunoglobulin G enzyme-linked immunosorbent assays were used to categorise individuals as seropositive or seronegative. H. pylori seropositivity and ANA positivity were common in the adult US population, with estimated prevalences of 33.3% and 9.9%, respectively. Both were associated with increasing age. H. pylori seropositivity was associated with higher odds of ANA (prevalence odds ratio = 1.89, 95% confidence interval = 1.08-3.33), adjusted for age, sex, race/ethnicity, educational attainment and body mass index. H. pylori infection may be one key factor in the loss of self-tolerance, contributing to immune dysfunction.

Pre-conception 25-hydroxyvitamin D (25(OH)D) and fecundability.

STUDY QUESTION: Is pre-conception 25(OH)D associated with the per cycle probability of conception, i.e fecundability, in a prospective cohort study? SUMMARY ANSWER: There are suggestive associations of high 25(OH)D (at least 50 ng/ml) with increased fecundability and low 25(OH)D (<20 ng/ml) with reduced fecundability, but the estimates were imprecise. WHAT IS KNOWN ALREADY: Vitamin D has been associated with reproductive function and fertility in animal studies, but few human studies exist. STUDY DESIGN, SIZE, DURATION: This community-based prospective cohort study included 522 women attempting to become pregnant between 2010 and 2016. The women completed online daily and monthly diaries until a positive home pregnancy test was observed or 12 months had elapsed. PARTICIPANTS/MATERIALS, SETTING, METHODS: The study included women from central North Carolina who were aged 30-44 with no history of infertility, with no more than 3 months of attempt time at recruitment. Women recorded vaginal bleeding so that the ongoing number of attempt cycles could be counted and used to quantify a woman's pregnancy attempt time. Blood collected at the study entry was analysed for 25(OH)D using liquid chromatography tandem mass spectrometry. Associations with fecundability were estimated with a log-binomial discrete time-to-event model. MAIN RESULTS AND THE ROLE OF CHANCE: Among 522 women, 257 conceived during the study. The mean age was 33 years and the mean 25(OH)D was 36 ng/ml. There was an estimated 10% higher fecundability with each 10 ng/ml increase in 25(OH)D (fecundability ratio (FR) 1.10, 95% CI: 0.96, 1.25). The suggestive dose-response association with the continuous measure of 25(OH)D was driven by women in the lowest and the highest categories of 25(OH)D. Compared to women with 25(OH)D of 30-40 ng/ml, women below 20 ng/ml had an estimated 45% reduction in fecundability (FR (CI): 0.55 (0.23, 1.32)), and women with at least 50 ng/ml had an estimated 35% increase in fecundability (FR (CI): 1.35 (0.95, 1.91)). Across these three categories (25(OH)D of <20 ng/ml, 30-40 ng/ml and > 50 ng/ml), the probability of taking longer than 6 months to conceive was, respectively, 51% (17%, 74%), 28% (17%, 39%) and 15% (10%, 37%). LIMITATIONS, REASONS FOR CAUTION: While the distribution of 25(OH)D was wide, the number of observed cycles with high 25(OH)D (N = 107) or low 25(OH)D (N = 56) was small. WIDER IMPLICATIONS OF THE FINDINGS: Our findings are consistent with prior reports of reduced fertility in women with 25(OH)D concentrations below the clinically defined deficiency level (20 ng/ml). Further studies are needed to evaluate the possible reproductive benefits of considerably higher 25(OH)D concentration (>50 ng/ml). STUDY FUNDING/COMPETING INTEREST(S): This research was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health (NIH) under award numbers R00HD079659 and R01HD067683 and supported in part by the Intramural Research Program of the National Institute of Environmental Health Sciences, under projects ES103086, ES049003 and ES044003. ClearBlue ovulation predictor kits were generously donated to AMZJ and AJW by Swiss Precision Diagnostics. Drs Wilcox and Jukic report non-financial support from Swiss Precision Diagnostics during the conduct of the study; Dr Jukic reports non-financial support from Theralogix, LLC, outside the submitted work. Otherwise there are no competing interests. TRIAL REGISTRATION NUMBER: N/A.

Simulating autosomal genotypes with realistic linkage disequilibrium and a spiked-in genetic effect.

BACKGROUND: To evaluate statistical methods for genome-wide genetic analyses, one needs to be able to simulate realistic genotypes. We here describe a method, applicable to a broad range of association study designs, that can simulate autosome-wide single-nucleotide polymorphism data with realistic linkage disequilibrium and with spiked in, user-specified, single or multi-SNP causal effects. RESULTS: Our construction uses existing genome-wide association data from unrelated case-parent triads, augmented by including a hypothetical complement triad for each triad (same parents but with a hypothetical offspring who carries the non-transmitted parental alleles). We assign offspring qualitative or quantitative traits probabilistically through a specified risk model and show that our approach destroys the risk signals from the original data. Our method can simulate genetically homogeneous or stratified populations and can simulate case-parents studies, case-control studies, case-only studies, or studies of quantitative traits. We show that allele frequencies and linkage disequilibrium structure in the original genome-wide association sample are preserved in the simulated data. We have implemented our method in an R package (TriadSim) which is freely available at the comprehensive R archive network. CONCLUSION: We have proposed a method for simulating genome-wide SNP data with realistic linkage disequilibrium. Our method will be useful for developing statistical methods for studying genetic associations, including higher order effects like epistasis and gene by environment interactions.

Maternal age at birth and daughters' subsequent childlessness.

STUDY QUESTION: Does maternal age at a daughter's birth predict her subsequent probability of lifelong childlessness? SUMMARY ANSWER: In this study population, women born to older mothers were more likely to be childless. WHAT IS KNOWN ALREADY: Although maternal age at childbearing is increasing in many countries, there is limited evidence on whether being born to older parents may influence offspring fertility. STUDY DESIGN SIZE AND DURATION: This analysis included 43 135 women from the US-based Sister Study, a cohort study of 50 884 sisters of women with breast cancer recruited between 2003 and 2009. PARTICIPANTS/MATERIALS, SETTING, METHODS: Participants had no breast cancer at baseline. Women were included in the analytic sample if they were born between 1930 and 1964 and were at least 44 years old at enrolment. Median age when reproductive history was last ascertained was 63.8 years. We estimated relative risks (RR) and 95% CI of lifelong childlessness as a function of maternal age at birth, using multivariable log-binomial models, including total number of siblings, birth order, socioeconomic indicators of the family of origin, race and birth cohort. We examined the association in different subgroups and in a sibling-matched analysis including 802 sister pairs discordant for childlessness. MAIN RESULTS AND ROLE OF CHANCE: Compared with women born to 20-24-year-old mothers, those born to mothers aged 25-29, 30-34 and ≥35 years were more likely to be childless [RR (95% CI): 1.21 (1.14-1.29), 1.30 (1.22-1.39) and 1.40 (1.31-1.50), respectively]. The association was consistent in strata defined by birth cohort, number of siblings, birth order, and participant's educational level, as well as within sister pairs. Overall, we found weak evidence for an independent contribution of paternal age at birth to the daughter's probability of childlessness. LIMITATIONS REASONS FOR CAUTION: All participants had at least one sister, and all information was self-reported. We had no knowledge of whether childlessness was intentional and found only a modest association between maternal age at birth and self-reported indicators of infertility. Still, the association with childlessness was highly consistent. WIDER IMPLICATIONS OF THE FINDING: Given the widespread tendency to delay childbearing, evaluating the influence of maternal age at birth on offspring fertility is a public health priority. STUDY FUNDING/COMPETING INTERESTS: This research was supported in part by the Intramural Research Programme of the NIH, National Institute of Environmental Health Sciences (Z01-ES044005). The authors report no conflict of interest.

Invited Commentary: Self-Control Is a Virtue.

The case-crossover design, introduced in 1991 by Malcolm Maclure (Am J Epidemiol. 1991;133(2):144-153), provided a precise and powerful tool for studying short-term effects of transient triggering exposures on abrupt outcomes like myocardial infarction. The design is an example of "self-control." One compares case-time intervals that include experiences that came just before the health event with control-time intervals that capture comparable experiences more remote from the event. Methodologists have since tweaked the general approach, recognizing issues that need to be considered to guard against time-driven confounders. I discuss opportunities for possible expansion and further mining of the data from this ingenious design.

Addressing data privacy in matched studies via virtual pooling.

BACKGROUND: Data confidentiality and shared use of research data are two desirable but sometimes conflicting goals in research with multi-center studies and distributed data. While ideal for straightforward analysis, confidentiality restrictions forbid creation of a single dataset that includes covariate information of all participants. Current approaches such as aggregate data sharing, distributed regression, meta-analysis and score-based methods can have important limitations. METHODS: We propose a novel application of an existing epidemiologic tool, specimen pooling, to enable confidentiality-preserving analysis of data arising from a matched case-control, multi-center design. Instead of pooling specimens prior to assay, we apply the methodology to virtually pool (aggregate) covariates within nodes. Such virtual pooling retains most of the information used in an analysis with individual data and since individual participant data is not shared externally, within-node virtual pooling preserves data confidentiality. We show that aggregated covariate levels can be used in a conditional logistic regression model to estimate individual-level odds ratios of interest. RESULTS: The parameter estimates from the standard conditional logistic regression are compared to the estimates based on a conditional logistic regression model with aggregated data. The parameter estimates are shown to be similar to those without pooling and to have comparable standard errors and confidence interval coverage. CONCLUSIONS: Virtual data pooling can be used to maintain confidentiality of data from multi-center study and can be particularly useful in research with large-scale distributed data.

Length of human pregnancy and contributors to its natural variation.

STUDY QUESTION: How variable is the length of human pregnancy, and are early hormonal events related to gestational length? SUMMARY ANSWER: Among natural conceptions where the date of conception (ovulation) is known, the variation in pregnancy length spanned 37 days, even after excluding women with complications or preterm births. WHAT IS KNOWN ALREADY: Previous studies of length of gestation have either estimated gestational age by last menstrual period (LMP) or ultrasound (both imperfect measures) or included pregnancies conceived through assisted reproductive technology. STUDY DESIGN, SIZE, DURATION: The Early Pregnancy Study was a prospective cohort study (1982-85) that followed 130 singleton pregnancies from unassisted conception to birth, with detailed hormonal measurements through the conception cycle; 125 of these pregnancies were included in this analysis. PARTICIPANTS/MATERIALS, SETTING, METHODS: We calculated the length of gestation beginning at conception (ovulation) in 125 naturally conceived, singleton live births. Ovulation, implantation and corpus luteum (CL) rescue pattern were identified with urinary hormone measurements. We accounted for events that artificially shorten the natural length of gestation (Cesarean delivery or labor induction, i.e. 'censoring') using Kaplan-Meier curves and proportional hazards models. We examined hormonal and other factors in relation to length of gestation. We did not have ultrasound information to compare with our gold standard measure. MAIN RESULTS AND THE ROLE OF CHANCE: The median time from ovulation to birth was 268 days (38 weeks, 2 days). Even after excluding six preterm births, the gestational length range was 37 days. The coefficient of variation was higher when measured by LMP (4.9%) than by ovulation (3.7%), reflecting the variability of time of ovulation. Conceptions that took longer to implant also took longer from implantation to delivery (P = 0.02). CL rescue pattern (reflecting ovarian response to implantation) was predictive (P = 0.006): pregnancies with a rapid progesterone rise were longer than those with delayed rise (a 12-day difference in the median gestational length). Mothers with longer gestations were older (P = 0.02), had longer pregnancies in other births (P < 0.0001) and were heavier at birth (P = 0.01). We did not see an association between the length of gestation and several factors that have been associated with gestational length in previous studies: body mass index, alcohol intake, parity or offspring sex. LIMITATIONS, REASONS FOR CAUTION: The sample size was small and some exposures were rare, reducing power to detect weak associations. WIDER IMPLICATIONS OF THE FINDINGS: Human gestational length varies considerably even when measured exactly (from ovulation). An individual woman's deliveries tend to occur at similar gestational ages. Events in the first 2 weeks after conception are predictive of subsequent pregnancy length, and may suggest pathways underlying the timing of delivery. STUDY FUNDING/COMPETING INTEREST: This research was supported by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences. None of the authors has any conflict of interest to declare.

The association of maternal factors with delayed implantation and the initial rise of urinary human chorionic gonadotrophin.

BACKGROUND Late implantation and the pattern of early rise in hCG have been associated with early pregnancy loss. We explored factors that might be predictive of these markers of poor embryonic health in spontaneously conceived pregnancies. METHODS Participants in the North Carolina Early Pregnancy Study collected daily first-morning urine specimens while attempting to conceive. Samples were assayed for estrogen and progesterone metabolites (to identify day of ovulation) and hCG (to detect conception). Data were available for 190 pregnancies, 48 of which ended in early loss (within 6 weeks of the last menstrual period). We used logistic regression to identify characteristics associated with late implantation (≥10 days post-ovulation). For pregnancies surviving at least 6 weeks (n= 142), we used linear mixed models to identify factors associated with variations in hCG rise in the first 7 days from detection. RESULTS Later implantation was associated with current maternal smoking [odds ratio (OR): 5.7; 95% confidence interval (CI): 1.1-30] and with oocytes that were likely to have been fertilized late in their post-ovulatory lifespan (OR: 5.1; CI: 1.9-16). Older women had a faster rise in hCG (P= 0.01), as did women who had relatively late menarche (P for trend = 0.02). Women exposed in utero to diethylstilbestrol showed an unusual pattern of slow initial hCG rise followed by a fast increase, a pattern significantly different from that of unexposed women (P= 0.002). CONCLUSIONS Although limited by small numbers and infrequent exposures, our analyses suggest that a woman's exposures both early in life and at the time of pregnancy may influence early development of the conceptus.

Less is more, except when less is less: Studying joint effects.

Most diseases are complex in that they are caused by the joint action of multiple factors, both genetic and environmental. Over the past few decades, the mathematical convenience of logistic regression has served to enshrine the multiplicative model, to the point where many epidemiologists believe that departure from additivity on a log scale implies that two factors interact in causing disease. Other terminology in epidemiology, where students are told that inequality of relative risks across levels of a second factor should be seen as "effect modification," reinforces an uncritical acceptance of multiplicative joint effect as the biologically meaningful no-interaction null. Our first task, when studying joint effects, is to understand the limitations of our definitions for "interaction," and recognize that what statisticians mean and what biologists might want to mean by interaction may not coincide. Joint effects are notoriously hard to identify and characterize, even when asking a simple and unsatisfying question, like whether two effects are log-additive. The rule of thumb for such efforts is that a factor-of-four sample size is needed, compared with that needed to demonstrate main effects of either genes or exposures. So strategies have been devised that focus on the most informative individuals, either through risk-based sampling for a cohort, or case-control sampling, extreme phenotype sampling, pooling, two-stage sampling, exposed-only, or case-only designs. These designs gain efficiency, but at a cost of flexibility in models for joint effects. A relatively new approach avoids population controls by genotyping case-parent triads. Because it requires parents, the method works best for diseases with onset early in life. With this design, the role of autosomal genetic variants is assessed by in effect treating the nontransmitted parental alleles as controls for affected offspring. Despite advantages for looking at genetic effects, the triad design faces limitations when examining joint effects of genetic and environmental factors. Because population-based controls are not included, main effects for exposures cannot be estimated, and consequently one only has access to inference related to a multiplicative null. We have proposed a hybrid approach that offers the best features of both case-parent and case-control designs. Through genotyping of parents of population-based controls and assuming Mendelian transmission, power is markedly enhanced. One can also estimate main effects for exposures and now flexibly assess models for joint effects.

Interactions between a Polygenic Risk Score and Non-genetic Risk Factors in Young-Onset Breast Cancer.

Most gene-environmental studies have focused on breast cancers generally, the preponderance of which occur after age 50. Young-onset breast cancers (YOBC) tend to be aggressive and may be etiologically different. The goal of this analysis was to assess interactions between an established 77-SNP polygenic risk score (PRS) and non-genetic risk factors for YOBC. We constructed the PRS using a family-based study of 1,291 women diagnosed with breast cancer before age 50 and their parents and unaffected sisters. We used conditional logistic regression to analyze interactions between the PRS and 14 established risk factors. In further analyses we assessed the same interactions, but for invasive cancer, estrogen receptor (ER) positive cancer and with broader inclusion of racial/ethnic groups. Results showed a decreased association between the PRS and YOBC risk for women who had ever used hormonal birth control (odds ratio [OR] = 2.20 versus 3.89) and a stronger association between the PRS and YOBC risk in pre-menopausal women (OR = 2.46 versus 1.23). Restricting the analysis to ER+ cancers or invasive cancers or using samples from all ethnic groups produced similar results. In conclusion, the PRS may interact with hormonal birth control use and with menopausal status on risk of YOBC.

Making the most of case-mother/control-mother studies.

The prenatal environment plays an important role in many conditions, particularly those with onset early in life, such as childhood cancers and birth defects. Because both maternal and fetal genotypes can influence risk, investigators sometimes use a case-mother/control-mother design, with mother-offspring pairs as the unit of analysis, to study genetic factors. Risk models should account for both the maternal genotype and the correlated fetal genotype to avoid confounding. The usual logistic regression analysis, however, fails to fully exploit the fact that these are mothers and offspring. Consider an autosomal, diallelic locus, which could be related to disease susceptibility either directly or through linkage with a polymorphic causal locus. Three nested levels of assumptions are often natural and plausible. The first level simply assumes Mendelian inheritance. The second further assumes parental mating symmetry for the studied locus in the source population. The third additionally assumes parental allelic exchangeability. Those assumptions imply certain nonlinear constraints; the authors enforce those constraints by using Poisson regression together with the expectation-maximization algorithm. Calculations reveal that improvements in efficiency over the usual logistic analysis can be substantial, even if only the Mendelian assumption is honored. Benefits are even more marked if, as is typical, information on genotype is missing for some individuals.

Urinary hCG patterns during the week following implantation.

BACKGROUND: Human chorionic gonadotrophin (hCG) is used to monitor pregnancy status. Yet the pattern of hCG excretion in the first week following implantation has not been adequately described. Therefore the aim of this study was to describe the average profile of hCG and its variability during the 7 days following estimated implantation in a population of naturally conceived pregnancies. METHODS: We measured daily hCG concentrations in first-morning urine for 142 clinical pregnancies from women with no known fertility problems. Mixed-effects regression models were used to estimate the hCG trajectory and its variability in relation to pregnancy outcomes. RESULTS: hCG rose 3-fold between the day of detection and the next day (95% CI = 2.7-3.4). The relative rate of rise decreased thereafter, reaching 1.6-fold (95% CI = 1.5-1.8) between days 6 and 7. HCG levels followed a log-quadratic trajectory, and the patterns of rise were unrelated to number of fetuses, risk of spontaneous abortion or sex of the baby. Later implantations (after 10 luteal days) produced slower rates of increase. CONCLUSIONS: Although mean hCG follows a log-quadratic trajectory during the first week of detectability, there is high variability across pregnancies. Later implantation may reflect characteristics of the uterus or conceptus that slow hCG production.

Current serum levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin in phenoxy acid herbicide applicators and characterization of historical levels.

BACKGROUND: Workers who sprayed phenoxy acid herbicides, especially those who sprayed before 1975, may have been exposed to significant amounts of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), a potent animal carcinogen present in herbicide preparations as a contaminant. PURPOSE: The aims of this study were (a) to determine serum levels of TCDD in a representative sample of workers occupationally exposed to the agent during the spraying of phenoxy acid herbicides; (b) to compare serum levels in workers exposed before 1965, when concentrations in herbicide products were unregulated and high, with levels in workers exposed after 1974, when concentrations were lower as a result of government regulations worldwide; and (c) to examine the correlation, if any, between serum levels and duration of employment in spraying. METHODS: Thirty-seven subjects were randomly selected from a group of 654 men who had sprayed the herbicides 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and 2,4-dichlorophenoxyacetic acid (2,4-D) in Australia for at least 12 months. The workers were classified as follows: eight who sprayed only before 1965, nine who sprayed only during the period after 1964 and before 1975, and 20 who sprayed during the period after 1974 and before 1991. Serum from the workers was analyzed for TCDD by high-resolution gas chromatography and high-resolution mass spectrometry at a detection limit of 0.6 parts per trillion (ppt) on a lipid-weight basis. In addition, rates of exposure to TCDD were estimated, as were TCDD serum concentrations at termination of employment and intensity of herbicide use. RESULTS: Only one worker, with a serum TCDD level of 34 ppt, had a serum level higher than the maximum level of 26 ppt reported for the general population. Assuming a half-life of 7.1 years, we estimated the mean exposure rates to be 2.7, 2.3, and 0.06 ppt/mo for the three epochs, respectively. We found the highest serum level of TCDD at the time of cessation of employment to be 329 ppt. Calendar period and intensity of use of 2,4,5-T and 2,4-D were statistically significant determinants of rate of exposure to TCDD, but 2,4-D was associated with exposure rate only for the pre-1975 periods. Estimated rates prior to 1965 were more than an order of magnitude higher than those after 1974. CONCLUSION: The highest estimated exposure rate was 20.7 ppt/mo, which suggests that some sprayers may have been exposed to levels comparable with those that produce cancer in laboratory animals.

Development of a predictive model for symptomatic neuropathy in diabetes.

In evaluating therapeutic interventions aimed at preventing diabetic neuropathy, choosing a suitable measure of neural function is difficult, partly because the relation between most available objective measures and the development of symptomatic neuropathy (SN) is unclear. Using data from 67 diabetic patients, we developed a linear logistic regression model to assess the relationship between SN and a set of neural measures, including dark-adapted pupil size; pupillary latency; heart rate; a measure of respiratory sinus arrhythmia (R); the Valsalva ratio; and conduction velocities for the peroneal, median motor, and median sensory nerves. Models allowed for possible effects related to age, sex, duration and type of diabetes, glycosylated hemoglobin, and adiposity. Thirty-two of the patients reported SN (autonomic and/or sensorimotor). The best-fitting model for predicting the presence of any SN included only heart rate, duration of disease, and R. Exclusion of duration (P less than .01), or heart rate (P less than .05), or R (P less than .001) significantly impaired the fit of the model. To evaluate the temporally predictive power of the model, nine of the asymptomatic patients were reinterviewed 2 yr later by the same interviewer, who was unaware of the results of the modeling. Four of five to whom the model had assigned high probability of symptoms had indeed developed SN during the follow-up period, whereas none of the four assigned low probability had developed SN (P less than .05). Thus it seems that a measure of respiratory sinus arrhythmia provides an index of neural function strongly related to SN, and our follow-up data suggest that diminished R can be used to predict the development of SN in diabetes.

Modelling multiple ovulation, fertilization, and embryo loss in human fertility studies.

Models of human fertility that incorporate information on timing of intercourse have assumed that a single ovum is released each menstrual cycle. These models are misspecified if two or more viable ova are sometimes released in a single cycle, which is known to occur in dizygotic twin pregnancies. In this paper, we propose a model for multiple ovulation in humans. We assume that the unobservable number of viable ova in each cycle follows a multinomial distribution. Successful fertilization of each ovum depends on the ability of the cycle to support a pregnancy and on the aggregate of a set of unobservable Bernoulli trials representing the fertilizing effects of intercourse on various days. Our model accommodates general covariate effects, allows for heterogeneity among couples, and accounts for a sterile subpopulation of couples. Information on early detection of pregnancy can be incorporated to estimate the probability of embryo loss. We outline a Markov chain Monte Carlo algorithm for estimation of the posterior distributions of the parameters. The methods are applied to data from a North Carolina pregnancy study, and applications to studies of assisted reproduction are described.

Sensitivity of R-R variation and Valsalva ratio in assessment of cardiovascular diabetic autonomic neuropathy.

R-R variation and the Valsalva ratio are commonly used to quantitatively assess diabetic autonomic neuropathy (DAN). To assess the sensitivity of these two measures to parasympathetic ablation, 12 nondiabetic subjects were tested before and after graded doses (0.3-4.0 mg i.v.) of atropine. R-R variation was significantly reduced at 0.7 mg, whereas Valsalva ratio was not significantly smaller until the 2.0-mg dose of atropine. R-R variation continued to become progressively smaller during the 0.85-, 1.0-, and 2.0-mg doses. Valsalva ratio, but not R-R variation, was further reduced by the 4.0-mg dose. To further compare these two measures, two groups of diabetic subjects were compared with a group of nondiabetic subjects (n = 22). One group of diabetic subjects had symptoms of DAN (n = 22), and the other diabetic group had no symptoms of DAN (n = 19). In DAN subjects, both R-R variation (nondiabetic 33.2 +/- 4.3 vs. DAN 9.8 +/- 1.2, P less than .001) and the Valsalva ratio (nondiabetic 1.98 +/- 0.07 vs. DAN 1.55 +/- 0.07, P less than .001) were reduced. However, in asymptomatic subjects, R-R variation (23.2 +/- 3.9, P less than .05), but not Valsalva ratio (1.94 +/- 0.13, NS), was less than nondiabetic subjects. Even after beta-blockade, R-R variation was still less in both groups of diabetic subjects (nondiabetic 34.4 +/- 4.2 vs. DAN 7.4 +/- 1.3, P less than .001; asymptomatic 21.8 +/- 3.3, P less than .02).(ABSTRACT TRUNCATED AT 250 WORDS)

Correlations among autonomic, sensory, and motor neural function tests in untreated non-insulin-dependent diabetic individuals.

A well-defined group of untreated non-insulin-dependent (NIDD) subjects were evaluated to determine whether involvement of neural function measurements is generalized and symmetrical and to compare the autonomic, sensory, and motor neural measurements. After age adjustment, the sensory and motor neural function measurements were significantly slower in the diabetic group than in normal subjects (P less than 0.01). Similarly, the autonomic nervous system function measurements were also abnormal in the NIDD group (P less than 0.01). Further analysis revealed that each of the specific measurements--median motor nerve conduction velocity (NCV,P less than 0.005), peroneal motor NCV (P less than 0.005), median sensory NCV (P less than 0.005), dark-adapted pupil size after muscarinic blockade (P less than 0.02), pupillary latency time (P less than 0.02), and RR-variation after beta adrenergic blockade (P less than 0.001)--was significantly less by analysis of covariance after age adjustment in the NIDD group than in normal subjects. Thus, there was evidence of motor and sensory neural impairment in the upper and lower extremities as well as evidence of impairment of the reflex arcs involving the parasympathetic nerves to the heart and eye and the sympathetic nerves to the iris. Further analysis revealed that right and left NCV were correlated (P less than 0.01), as were the median motor and median sensory NCV (P less than 0.01), the median motor and peroneal motor NCV (P less than 0.001), and the peroneal motor and median sensory NCV (P less than 0.001). Thus, there was evidence of symmetrical upper and lower limb, as well as motor and sensory proportional involvement of large nerve fiber NCV in this group of NIDD subjects.(ABSTRACT TRUNCATED AT 250 WORDS)