Heterogeneity in determinants of familial resemblance of lipid and lipoprotein levels between populations in North America and Israel was investigated using path analysis. A common protocol, identical measurement techniques, and the same statistical procedures were used in the two samples. Both genetic (h 2 ) and cultural (c 2 ) determinants of inheritance were significant for all lipid variables in the two studies. Genetic and cultural heritability of total cholesterol (h 2 = 0.61, c 2 = 0.02), low‐density lipoprotein cholesterol (h 2 = 0.59, c 2 = 0.02), and high‐density lipoprotein cholesterol (h 2 = 0.55, c 2 = 0.06) did not differ significantly between North America and Israel, while there was a significant difference for triglyceride (h 2 = 0.41, c 2 = 0.07 in North America; h 2 = 0.61, c 2 = 0.05 in Israel). Secondary parameters of the path model describing intrafamilial environmental relationships differed between the two countries. In particular, there was a higher correlation between marital environments in Israel for all traits except triglyceride, and a larger effect of father's environment on offspring's environment in Israel for all traits. Within both populations, variation of plasma lipids and lipoproteins was mostly explained by genetic factors and random unmeasured environmental factors. The contribution of common family environment was found to be small, though statistically significant. This is probably due to homogeneity of the distribution of familian environmental determinants within both countries.

Parent-offspring associations of total cholesterol and triglycerides were compared between family dyads in six North American populations examined between 1972 and 1976 as part of the North American Lipid Research Clinics Prevalence Study and those from families examined between 1976 and 1979 at the Lipid Research Clinic located in Jerusalem. Common study design, protocol, and laboratory techniques were used by all Lipid Research Clinics. The authors first examined homogeneity of familial correlations across clinics in the North American population and across origin groups in the Israeli sample. In general, correlations were homogeneous across clinics and origin groups, except for parent-daughter pairs for triglycerides in North America. The pooled familial correlations were similar in the two study populations. There was no asymmetry in parent-offspring correlations by the sex of the offspring. The pooled mother-child correlations were significantly higher than father-child values in the North American sample only. The strength of parent-offspring similarity showed no consistent pattern of change with level of education of parents in either study group. Patterns of familial similarity are discussed in relation to genetic, cultural, and environmental differences between the two study populations.

Complex segregation analysis with the unified mixed model in white families from nine lipid research clinics was carried out to delineate the mode of familial transmission of plasma high-density-lipoprotein cholesterol (HDL-C). Three groups of families from the collaborative Lipid Research Clinics Program Family Study were assessed: 1,146 selected at random, 483 obtained through hypercholesterolemic probands, and 177 selected from the random sample because a number had low HDL-C, the sample sizes being 4,279, 1,807 and 735, respectively. The data were first transformed and adjusted for effects of covariates. Analyses were performed within clinic and selection strata and also pooled across clinics within strata. The results were consistent across strata and identified two major HDL-C clusters with means separated by approximately 3 SD. There was significant evidence of transmission of a major factor for low HDL-C, but transmission did not conform to Mendelian segregation expectations. There was also evidence of significant multifactorial transmission. Since low HDL-C levels are a major independent risk factor for coronary heart disease, the association of a major factor with familial aggregation of low HDL-C emphasizes the importance of detailed within-family sampling for low HDL-C after identifying a proband whose predominant dyslipoproteinemia is low HDL-C.

The single major locus (SML) model of Bucher and Elston (1981) was applied to data collected in a long-term follow-up and family study of major effective disorders. Pedigree segregation analysis was used to test the hypothesis that bipolar and unipolar disorders are phenotypic variants of the same genotype at a SML. The particular SML model examined did not adequately describe the familial pattern of these disorders in families of bipolar and unipolar probands. Although other SML models may be superior, it is likely that the existence of a SML will be obscured by genetic heterogeneity.

Summary Family data from schizophrenic and control probands were analyzed using a descriptive analysis and multiple threshold models to determine whether a given group of diagnoses made in accordance with ICD–9 was aetiologically related to schizophrenia. The proportion of relatives receiving any psychiatric diagnosis, other than schizophrenia and affective disorder, was essentially the same between the two study groups. Furthermore, the data did not fit the multiple threshold model tested. Thus, the hypothesis that schizophrenia and a spectrum of disorders defined according to ICD–9 have a common familial aetiology was not accepted.

Summary Segregation analysis was applied to blind family data concerning schizophrenia to decide if the transmission of schizophrenia could be explained by a single major gene. Our results showed that the Mendelian model was unacceptable. Therefore, the monogenic hypothesis could not account for the transmission of schizophrenia. Since the hypothesis of no parent-child transmission was also not accepted, there was an indication that some form of vertical transmission existed which could be psychosocial, or an interaction between genetic and psychosocial factors. Our results suggest genetic heterogeneity in schizophrenia. Currently available clinical criteria for defining subgroups must be improved in conjunction with detection of biological indicators so that segregation analysis of family data could be effectively used in determining modes of transmission in schizophrenia.

The Muscatine Cholesterol Family Study includes data on first, second and third-degree relatives of school children who have the following cholesterol levels: above the 95th percentile in two consecutive school screens, N = 53; below the 5th percentile twice, N = 45; and a random sample between the 5th and 95th percentiles twice, N = 44. For males and females not taking estrogen the relationship of cholesterol to age was the same in high, middle, and low groups, although the overall level differed. For females taking estrogen, cholesterol levels rose more quickly with age in the high group as compared to the middle and low groups. The triglyceride levels of females taking estrogen is higher at younger ages and shows no significant change with increasing age. Among first degree relatives, the cholesterol correlations are between 0.25 and 0.30. Correlations in relatives of the middle group probands are higher than those in relatives of high or low group probands, but not significantly so. Correction for ascertainment increases the correlations in the high and low groups. Among second degree relatives cholesterol correlations in all three groups are approximately 0.20. There is negligible cholesterol correlation between spouses in the high and low groups, while the middle group spouse correlation is significantly greater than zero. Categorical analysis suggests the effect of the proband's cholesterol level is seen most upon the cholesterol level of parents, with siblings and second-degree relatives being less similarly affected.

The Muscatine Cholesterol Family Study includes the relatives of three groups of schoolchildren: those with cholesterol levels above the 95th percentiles in two consecutive school screens, those with cholesterol levels below the 5th percentile twice, and random sample of those with cholesterol levels between the 5th and 95th percentiles twice. This paper examines the cholesterol distribution within and among these families for evidence of major gene effects. Three screening techniques are used: admixture analysis using maximum likelihood, comparison of within-sibship variances among groups, and regression of within-sibships variance on within-sibship mean. The results of the three techniques are consistent and indicate the existence of a major gene in a subset of the families with a proband above the the 95th percentile, but not in the other two groups of families. We estimate that 15% of schoolchildren with a cholesterol level above the 95th percentile twice, or 3 per thousand in the general population, have a dominant gene as a cause of the cholesterol elevation.

Pedigree analysis is done on 19 kindreds of panic disorder, and the results suggest that this disorder is transmitted as an autosomal dominant trait. Seven of these 19 kindreds were ascertained through a panic disorder proband with mitral valve prolapse. When the analysis is done omitting these seven kindreds, the results also suggest that panic disorder without prolapse is transmitted as an autosomal dominant trait.

64 families containing a proband with corneal endothelial dystrophy were examined in order to study the hereditary nature of the disease. Data concerning the frequency of occurrence, severity of the disease, ratio of affected females to males, relationship of the disease with age, and other factors were the subject of a previous report. 7 pedigrees which reflect features of endothelial dystrophy within the 64 families are presented. These features include multiple females in a family being affected, multiple consecutively affected generations, the occurrence of offspring with disease more severe than the parent, and endothelial decompensation (edema) at a relatively young age (less than 40 years of age). The importance of examining family members whenever possible rather than relying on history alone is emphasized. A statistical analysis of the inheritance pattern was performed. Endothelial dystrophy does not seem to follow a strict autosomal dominant pattern even though superficial inspection suggests autosomal dominant inheritance (both males and females affected, successive generations affected, 38% of relatives over the age of 40 years affected). Even though we were unable to determine a specific genetic mode of inheritance in these 64 families with endothelial dystrophy, we do feel that endothelial dystrophy is at least in part an inherited disease. Future investigations might prove sex-linked dominance, genetic heterogeneity, the influence of environmental factors, or a multifactorial etiology.

A prospective study was undertaken during an 18-month period with 64 families who had endothelial dystrophy. Two hundred twenty-eight relatives were examined. Of those older than the age of 40, 38% were affected. Women were affected more severely and 2.5 times more frequently than men. The disease showed a strong familial tendency: there was one family in which three generations were affected and 16 families in which two generations were affected. There were four families that had members with edema in two generations. There was no association between edema in a parent and edema in a child. The proportion of relatives affected and the severity of involvement increased with age. Fifty-three probands and 18 relatives had endothelial dystrophy with edema (Fuchs' dystrophy). Of these 71, one had glaucoma.