A simulation approach for power calculation in large cohort studies based on multistate models

Realistic power calculations for large cohort studies and nested case control studies are essential for successfully answering important and complex research questions in epidemiology and clinical medicine. For this, we provide a methodical framework for general realistic power calculations via simulations that we put into practice by means of an R‐based template. We consider staggered recruitment and individual hazard rates, competing risks, interaction effects, and the misclassification of covariates. The study cohort is assembled with respect to given age‐, gender‐, and community distributions. Nested case‐control analyses with a varying number of controls enable comparisons of power with a full cohort analysis. Time‐to‐event generation under competing risks, including delayed study‐entry times, is realized on the basis of a six‐state Markov model. Incidence rates, prevalence of risk factors and prefixed hazard ratios allow for the assignment of age‐dependent transition rates given in the form of Cox models. These provide the basis for a central simulation‐algorithm, which is used for the generation of sample paths of the underlying time‐inhomogeneous Markov processes. With the inclusion of frailty terms into the Cox models the Markov property is specifically biased. An “individual Markov process given frailty” creates some unobserved heterogeneity between individuals. Different left‐truncation‐ and right‐censoring patterns call for the use of Cox models for data analysis. p‐values are recorded over repeated simulation runs to allow for the desired power calculations. For illustration, we consider scenarios with a “testing” character as well as realistic scenarios. This enables the validation of a correct implementation of theoretical concepts and concrete sample size recommendations against an actual epidemiological background, here given with possible substudy designs within the German National Cohort.     

Electrolocation in the presence of jamming signals: behavior

Electrolocation behavior of Apteronotus leptorhynchus was studied by monitoring the animal's ability to maintain orientation to a variety of moving electrolocation targets. The primary goal of this study was to determine the relative effectiveness of various types of electrical 'jamming signals' in disrupting electrolocation performance. 1. Two measures of electrolocation performance were used: The latency between the electrolocation target motion and the fish's following response, and the minimum distance separating the fish from the target during the target movement sequence. Latency increased and minimum fish-target distance decreased as target size was decreased, and when large diameter ceramic targets were used as control stimuli the fish were less able to avoid, and frequently collided with, these 'electrically transparent' objects. 2. Four types of jamming signals were used in attempts to mask the electrosensory input used for electrolocation. Broad-band noise and sinusoidal signals, different in frequency by a few Hz from the animal's personal electric organ discharge (DF stimuli), were used to jam the tuberous electroreceptors. Five Hz and 50 Hz sinusoidal signals were used to jam the low-frequency or ampullary receptor system. Both the noise and the DF stimuli were effective in reducing electrolocation performance, and the threshold intensity for behavior deterioration was about three-fold lower for DF stimuli as compared to the noise. The rate of change of response deterioration as a function of increasing jamming intensity was, however, not different for these two types of stimuli. Neither the 50 Hz nor the 5 Hz jamming signals caused electrolocation deterioration when presented alone. However, 5 Hz jamming, when added to either the noise or DF jamming, did result in significant increments in response deterioration. This suggests that the ampullary receptors can provide supplementary information for electrolocation. 3. Electrolocation performance deterioration was also studied with various difference frequencies between an animal's EOD and the sinusoidal jamming stimulus. Increasing DF results in decreased electrolocation deterioration, but even at the highest DF frequencies used (128 Hz) significant response degradation was observed. 4. The apparent differences in the effectiveness of noise and DF stimulation in reducing electrolocation performance are largely accounted for by the differential effects of the tuberous electroreceptor filter characteristics on these two types of signals.     

Microcephaly: general considerations and aids to nosology

Microcephaly is defined as an occipito-frontal head circumference (OFC) 2 or more standard deviations below the mean for age and sex using the new Roche et al. [Pediatrics 1987;79:706-712] charts, and corrected for parental OFC by the method of Weaver and Christian [J Pediatr 1980;96:990-994]. "Relative" microcephaly, i.e., a small head on a small child, may be associated with a much better intellectual prognosis than absolute microcephaly, although the average IQ of children with absolute microcephaly ascertained in a normal school system is normal when compared with that of appropriate control children. "Primary" microcephaly means an abnormal OFC at birth (corrected for gestational age and length), and "secondary" microcephaly a normal birth OFC with later, acquired microcephaly due to deceleration of brain growth reflecting infection, trauma, intoxication, metabolic disease, the Rett syndrome, or a true CNS degenerative disease. Some cases of syndromal microcephaly may be associated with normal intelligence including some "primordial dwarfs," children with Dubowitz syndrome, FAS, mild SC-Roberts syndrome, and an occasional Brachmann-de Lange individual. The nosology of (syndromal) microcephaly is extraordinarily complex and requires the assistance of special library resources and information retrieval expertise. At a minimum, it requires McKusick's Catalog of Mendelian Inheritance in Man (MIM); however, we find that our work is greatly enhanced by recently developed electronic databases such as MIM-online (OMIM), POSSUM, SYNDROME, and MEDLINE, as well. Three groups of syndromal and non-syndromal microcephaly are discussed selectively in order to illustrate the marvels of pleiotropy in human development and its abnormalities and the difficulties encountered in splitting and lumping entities with overlapping manifestations.     

Lymphocyte ecto-5'-nucleotidase activity in infancy: increasing activity in peripheral blood B cells precedes their ability to synthesize IgG in vitro

Ecto-5'-nucleotidase activity was measured in peripheral blood lymphocytes isolated from serial specimens from nine healthy full-term infants and two premature infants at 0, 2, 4, and 6 mo of age. The postnatal nadir in activity was 7.1 +/- 2.0 nmol/hr/10(6) cells, which is the same as the activity in cord blood lymphocytes (7.0 +/- 2 nmol/hr/10(6) cells). The activity rose twofold to 13.2 +/- 3.8 nmol/hr/10(6) cells at 6 mo of age (p less than 0.001, paired t-test), which is similar to the activity in adult peripheral blood lymphocytes (14.1 +/- 6.3 nmol/hr/10(6) cells). This increased activity in total lymphocytes reflects increased activity in the B cell population. B cell ecto-5'-nucleotidase activity in two infants at 12 to 13 mo of age was 19.3 and 25.2 nmol/hr/10(6) cells, values that are four-to fivefold higher than for cord blood B cells (5.6 +/- 2.8 nmol/hr/10(6) cells) and within the normal range for adult B cells (27.9 +/- 12 nmol/hr/10(6) cells). In spite of a greatly expanded peripheral blood B cell population, studies of immunoglobulin biosynthesis in vitro demonstrated that infant peripheral blood B cells are functionally immature with no synthesis of IgG in response to Epstein Barr virus. Thus, the increase in peripheral blood B lymphocyte ecto-5'-nucleotidase activity in infants precedes their acquisition of a capacity for IgG synthesis in vitro. Data from a hypogammaglobulinemic infant revealed a persistently low ecto-5'-nucleotidase activity over a 10-mo period until at 14 mo of age the activity was 8.8 nmol/hr/10(6) cells in total lymphocytes and 13.0 nmol/hr/10(6) cells in B cells, which correlated with in vivo and in vitro evidence of delayed B cell maturation. Thus, ecto-5'-nucleotidase activity may be a useful cell surface marker in studies of human postnatal B cell maturation.     

Phenotype and function of engrafted maternal T cells in patients with severe combined immunodeficiency

Engrafted maternal T cells from two patients with severe combined immunodeficiency (SCID) and graft-vs-host disease (GVHD) were characterized for surface phenotype, function, and ecto-5'-nucleotidase (ecto-5'-NT) activity. The majority of engrafted T cells from both patients were T6-, T3+, and Ia+; the ratio of T4+:T8+ cells varied from 0.89 to 3.1 for Patient 1 and was 0.17 for Patient 2. The sum of T4+ + T8+ cells was greater than the number of T3+ cells, and approximately one-third of the patients' T cells were T3-. Two-color immunofluorescent staining showed that one-third of the T cells from Patient 1 had a novel cell surface phenotype (T6-, T3-, T4+, T8+) that was not previously described. T cells from Patient 1 failed to proliferate in response to allogeneic cells or specific antigen and provided little help for PWM-driven Ig synthesis in vitro. However, they did suppress Ig synthesis in vitro and proliferate in response to PHA and Con A; thus they appeared to be more mature than the T cells of Patient 2 and of most previously reported patients with SCID and maternal T cell grafts. Both patients lacked detectable lymphocyte ecto-5'-NT activity, suggesting that either the ecto-5'-NT activity of maternal T cells is lost after engraftment or that a specific subset(s) of ecto-5'-NT-negative maternal T cells predominates in infants with SCID and GVHD. Thus, in vitro T cell function and the proportions of T cells bearing T4 and T8 may vary in SCID patients with maternal T cell grafts. However, the presence of the Ia antigen and the absence of ecto-5'-NT activity may be consistent features of activated maternal T cells responsible for GVHD.     

The physiology and morphology of two types of electrosensory neurons in the weakly electric fishApteronotus leptorhynchus

Summary Previous anatomical and physiological studies of the gymnotoid electrosensory lateral line lobe (ELLL) suggest that the anatomically identified basilar and non-basilar pyramidal cells correspond to the physiologically defined E and I cells. Intracellular injection of horseradish peroxidase (HRP) into physiologically identified E and I cells confirms this hypothesis. The morphologies and physiological responses of the basilar and non-basilar pyramidal cells were compared. Both types of pyramidal cells have extensive apical dendritic trees that interact with a parallel fiber network in the ELLL. The apical dendritic trees of the non-basilar pyramidal cells have a wider spread along the rostrocaudal axis of the ELLL than those of the basilar pyramidal cells. This difference is discussed in reference to the interaction of these cell types with the parallel fibers of the ELLL. The density of apical dendritic branches was measured and related to the distance of these branches from the cell body. No obvious differences were seen between the dendritic density patterns of basilar and non-basilar pyramidal cells. An interesting correlation, however, exists between the atypical physiological characteristics of two basilar pyramidal cells and their dendritic density patterns. Two cells of the medial (ampullary) segment of the ELLL were analyzed. Like the pyramidal cells of the three lateral (tuberous) regions of the ELLL, the physiology of these cells appears to be related to the presence of an extended basilar process. The ampullary cells, however, have apical dendritic trees that are oriented orthogonally to the dendritic trees of the pyramidal cells.Abbreviations AM amplitude modulation - DML dorsal molecular layer - ELLL electrosensory lateral line lobe - EOD electric organ discharge - HRP horseradish peroxidase - LC lobus caudalis - Npd nucleus praeeminentialis dorsalis - PSTH post stimulus time histogram