In a historical cohort study of 300,000 19-year-old men exposed to the Dutch famine of 1944-45 and examined at military induction, we tested the hypothesis that prenatal and early postnatal nutrition determines subsequent obesity. Outcomes were opposite depending on the time of exposure. During the last trimester of pregnancy and the first months of life, exposure produced significantly lower obesity rates (P less than 0.005). This result is consistent with the inference that nutritional deprivation affected a critical period of development for adipose-tissue cellularity. During the first half of pregnancy, however, exposure resulted in significantly higher obesity rates (P less than 0.0005). This observation is consistent with the inference that nutritional deprivation affected the differentiation of hypothalamic centers regulating food intake and growth, and that subsequent increased food availability produced an accumulation of excess fat in an organism growing to its predetermined maximum size.

Part I of this paper traced the evolution of modern epidemiology in terms of three eras, each with its dominant paradigm, culminating in the present era of chronic disease epidemiology with its paradigm, the black box. This paper sees the close of the present era and foresees a new era of eco-epidemiology in which the deployment of a different paradigm will be crucial. Here a paradigm is advocated for the emergent era. Encompassing many levels of organization--molecular and societal as well as individual--this paradigm, termed Chinese boxes, aims to integrate more than a single level in design, analysis, and interpretation. Such a paradigm could sustain and refine a public health-oriented epidemiology. But preventing a decline of creative epidemiology in this new era will require more than a cogent scientific paradigm. Attention will have to be paid to the social processes that foster a cohesive and humane discipline.

In multivariate analyses, a number of specific measures of respiratory care practice during the first postnatal week were associated with the risk of a very low birth weight infant developing CLD. However, after adjusting for baseline risk, most of the increased risk of CLD among very low birth weight infants hospitalized at 2 Boston NICUs, compared with those at Babies' Hospital, was explained simply by the initiation of mechanical ventilation.

In this paper, criteria used by many epidemiologists as aids in causal inference are reviewed and revised. The revised scheme emphasizes the distinction between essential properties of a cause and criteria useful for deciding on the presence of these properties in a given case. A systematic procedure for causal inference tests each essential causal property in turn against appropriate criteria. For a pragmatic epidemiology in which all determinants serve as causes, their essential properties are held to be association, time order, and direction, in an ascending hierarchy. Criteria for association are probabilistic and can be enhanced by strength and consistency. Given association, criteria for time order of the relevant variables follow from access to observation, which is dependent on design. Given association and time order, causal direction (or consequential change) calls on an array of criteria, namely, consistency and survivability, strength, specificity in cause and in effect, predictive performance, and coherence in all its forms (e.g., theoretical, factual, biologic, and statistical). The evolution of such criteria is traced through the epidemiologic literature in the light of historical context. Although Popper's philosophy cannot directly serve an inherently inductive judgmental process, his notion of survivability has here been added, alongside replicability, as a subclass of consistency. This criterion is proposed to bridge the gap between the particularity of designs and the generality required of causal relations. Designs are ordered and described in the framework of testing survivability. Finally, definitions are offered for the list of criteria deployed.

There is strong evidence for a genetic contribution to epilepsy, but it is commonly assumed that this genetic contribution is limited to 'generalized' epilepsies, and that most forms of 'partial' epilepsy are nongenetic. In a linkage analysis of a single family containing 11 affected individuals, we obtained strong evidence for localization of a gene for partial epilepsy. This susceptibility gene maps to chromosome 10q, with a maximum two-point lod score for D10S192 of 3.99 at theta = 0.0. All affected individuals share a single haplotype for seven tightly linked contiguous markers; the maximum lod score for this haplotype is 4.83 at theta = 0.0. Key recombinants place the susceptibility locus within a 10 centimorgan interval.