Biomedical subjects

M A Province

Publications and source records attributed to M A Province.

At least 19 recordsLinked to original sources

PubMed2001

The significance of not finding a gene.

As more investigators conduct extensive whole-genome linkage scans for complex traits, interest is growing in meta-analysis as a way of integrating the weak or conflicting evidence from multiple studies. However, there is a bias in the most commonly used meta-analysis linkage technique (i.e., Fisher's [1925] method of combining of P values) when it is applied to many nonparametric (i.e., model free) linkage results. The bias arises in those methods (e.g., variance components, affected sib pair, extremely discordant sib pairs, etc.) that truncate all "negative evidence against linkage" into the single value of LOD = 0. If incorrectly handled, this bias can artificially inflate or deflate the combined meta-analysis linkage results for any given locus. This is an especially troublesome problem in the context of a genome scan, since LOD = 0 is expected to occur over half the unlinked genome. The bias can be overcome (nearly) completely by simply interpreting LOD = 0 as a P value of 1divided by 2ln(2) is approximately equal to .72 in Fisher's formula.

Bias↗

PubMed2001

Usefulness of cardiovascular family history data for population-based preventive medicine and medical research (the Health Family Tree Study and the NHLBI Family Heart Study).

Detailed medical family history data have been proposed to be effective in identifying high-risk families for targeted intervention. With use of a validated and standardized quantitative family risk score (FRS), the degree of familial aggregation of coronary heart disease (CHD), stroke, hypertension, and diabetes was obtained from 122,155 Utah families and 6,578 Texas families in the large, population-based Health Family Tree Study, and 1,442 families in the NHLBI Family Heart Study in Massachusetts, Minnesota, North Carolina, and Utah. Utah families with a positive family history of CHD (FRS > or =0.5) represented only 14% of the general population but accounted for 72% of persons with early CHD (men before age 55 years, women before age 65 years) and 48% of CHD at all ages. For strokes, 11% of families with FRS > or =0.5 accounted for 86% of early strokes (<75 years) and 68% of all strokes. Analyses of >5,000 families sampled each year in Utah for 14 years demonstrated a gradual decrease in the frequency of a strong positive family history of CHD (-26%/decade) and stroke (-15%/decade) that paralleled a decrease in incidence rates (r = 0.86, p <0.001 for CHD; r = 0.66, p <0.01 for stroke). Because of the collaboration of schools, health departments, and medical schools, the Health Family Tree Study proved to be a highly cost-efficient method for identifying 17,064 CHD-prone families and 13,106 stroke-prone families (at a cost of about $27 per high-risk family) in whom well-established preventive measures can be encouraged. We conclude that most early cardiovascular events in a population occur in families with a positive family history of cardiovascular disease. Family history collection is a validated and relatively inexpensive tool for family-based preventive medicine and medical research.

Adult↗

PubMed2001

Tree-based recursive partitioning methods for subdividing sibpairs into relatively more homogeneous subgroups.

We propose a new splitting rule for recursively partitioning sibpair data into relatively more homogeneous subgroups. This strategy is designed to identify subgroups of sibpairs such that within-subgroup analyses result in increased power to detect linkage using Haseman-Elston regression. We assume that the subgroups can be defined by patterns of non-genetic binary covariates measured on each sibpair. The data we consider consists of the squared difference of a quantitative trait measurement on each sibpair, estimates of identity-by-descent (IBD) values at each genetic marker, and binary covariate data describing characteristics of the sibpair (e.g., race, sex, family history of disease). To test the efficacy of this method in linkage analysis, we performed two simulation experiments. In the first, we simulated a mixture consisting of 66.6% of the sibpairs with no linkage and 33.3% of the sibpairs with genetic linkage to one marker. The two groups were distinguished by the value of a single binary covariate. We also simulated one unlinked marker and one random covariate to include as noise in the data. In the second experiment, we simulated a mixture consisting of 55% of the sibpairs with no genetic linkage, 22.5% of the sibpairs with genetic linkage to one marker, and 22.5% of the sibpairs with linkage to a different marker. Each subgroup was defined by a distinct pattern of two binary covariates. We also simulated one unlinked marker and two random covariates to include as noise in the data. Our simulation studies found that we can significantly increase the overall power to detect linkage by fitting Haseman-Elston regression models to homogeneous subgroups with only a small increase in the false-positive rate. Second, the splitting rule can correctly identify important covariates and linked markers. Third, recursive partitioning of sibpair data using this splitting rule can correctly identify sibpair subgroups. These results indicate that partitioning sibpairs into homogeneous subgroups is feasible and significantly increases the power to detect linkage, thus demonstrating the practical utility and potential this new methodology holds.

Chromosome Mapping↗

PubMed2001

Linkage and association with structural relationships.

The use of structural equations (path analysis) provides an alternative, equivalent formulation to variance components models. Instead of partitioning the variance, we focus on modeling the underlying random variables themselves through a system of linear, mixed model, regression equations. A few specific examples of genetic path models for linkage and association (linkage disequilibrium) are discussed. This formulation provides a simple yet elegant framework that can continue to be extended to meet the challenges of modeling and dissecting the genetic nature of complex traits in the new century.

Chromosome Mapping↗

PubMed2001

Meta-analysis for model-free methods.

The intricate nature of complex genetic traits dictates that novel methodologies be developed and utilized to achieve better power, better accuracy, and more favorable balance between type I and type II errors than could be achieved by the traditional methods as they are used in mapping Mendelian traits. Meta-analysis provides one such method for synthesizing information from multiple studies. This has the advantage of being able to pool relatively weak signals from individual studies into a collectively stronger evidence of genetic effects, while at the same time providing a quantitative framework for modeling variability among studies. The traditional lod score measures significance level of a linkage effect in an individual study, and its additive property make it a natural candidate for combining results across independent studies. To incorporate the within-study variation of the linkage effect into the pooled overall measure of genetic effect, the effect sizes (such as the proportion of genes shared identical-by-descent, IBD) should be pooled directly across studies. Traditional regression models and mixed effects models can be used to estimate the overall genetic effect size and its variance, and to test heterogeneity among studies. Our simulation studies show that designing studies with moderate power and pooling their results via meta-analysis may be more cost-effective than large dedicated studies. We believe that, as a newly emerging methodology, the meta-analysis approach has the potential to become an integral part of our toolbox that will expedite the search for complex human disease genes.

Alleles↗

PubMed2001

Classification methods for confronting heterogeneity.

Recursive partitioning/tree models are discussed as a method of dissecting the complex nature of traits with different causal mechanisms operating in different subsets of the data (e.g., different genes operating in different subsets of families). In addition to the straightforward application of classification and regression trees to define more homogeneous subsets of the data on which to conduct further analysis, developments incorporating linkage analysis into the definition of the regression trees (Shannon et al., 2000) are discussed. The pros and cons of recursive partitioning vs. the related approach of context-dependent analysis (Turner et al., 1999) are also reviewed as two promising analysis strategies that may be useful for genetic dissection of complex traits.

Genetic Diseases, Inborn↗

PubMed2001

Sequential methods of analysis for genome scans.

As the preceding chapters illustrate, now that whole-genome scan analyses are becoming more common, there is considerable disagreement about the best way to balance between false positives and false negatives (traditionally called type I and type II errors in the statistical parlance). Type I and type II errors can be simultaneously controlled, if we are willing to let the sample size of analysis vary. This is the secret that Wald (1947) discovered in the 1940s that led to the theory of sequential sampling and was the inspiration for Newton Morton in developing the lod score method. We can exploit this idea further and capitalize on an old, but nearly forgotten theory: sequential multiple decision procedures (SMDP) (Bechhoffer, et al., 1968), which generalizes the standard "two-hypotheses" tests to consider multiple alternative hypotheses. Using this theory, we can develop a single, genome-wide test that simultaneously partitions all markers into "signal" and "noise" groups, with tight control over both type I and type II errors (Province, 2000). Conceiving this approach as an analysis tool for fixed sample designs (instead of a true sequential sampling scheme), we can let the data decide at which point we should move from the hypothesis generation phase of a genome scan (where multiple comparisons make the interpretation of p values and significance levels difficult and controversial), to a true hypothesis-testing phase (where the problem of multiple comparisons has been all but eliminated so that p values may be accepted at face value).

Chromosome Mapping↗

PubMed2001

Familial aggregation of QT-interval variability in a general population: results from the NHLBI Family Heart Study.

QT-interval prolongation is associated with increased risk of cardiac death. Although information on genetics and molecular mechanisms of the congenital long QT syndrome is mounting, limited data are available on the genetics of QT interval in the general population. Heart rate adjusted QT intervals (Bazett's QTc, and QT index (QTI)) were assessed by electrocardiography in 2399 members aged 25-91 years of 468 randomly selected families participating in the NHLBI Family Heart Study. Familial correlation and segregation analyses were performed to evaluate the genetics of the variability of QT interval in this population. The parent-offspring (0.14+/-0.03) and sibling (0.18+/-0.03) correlations for age and sex-adjusted QTc were moderate, while the spouse correlation was close to zero (0.09+/-0.06). This suggests that there are familial/genetic influences on QT-interval variability. Segregation analysis results suggest that there is a major effect in addition to heritable multifactorial effects (h2=0.34), but the major effect did not follow Mendelian inheritance. Further adjustments of QTc for other major cardiovascular risk factors did not significantly change the results. Similar results were found for QTI. The QT-interval variation in the general population is influenced by moderate heritable multifactorial effects in addition to a major effect. A major gene effect is not directly supported.

Adult↗

PubMed2001

Familial aggregation of submaximal aerobic performance in the HERITAGE Family study.

PURPOSE: This study examines the contribution of genetic factors to submaximal aerobic performance phenotypes measured before and after 20 wk of endurance training. METHODS: Submaximal oxygen consumption (VO(2)) at three power outputs, 50 W (VO(2)50W), 60% (VO(2)60%) and 80% (VO(2)80%) of VO(2max) and power outputs at 60% (PO60%) and 80% (PO80%) of VO(2max) were measured during cycle ergometer exercise tests in 483 subjects from 99 white families participating in the HERITAGE Family study. The baseline phenotypes were adjusted for the effects of age, sex, and body mass using stepwise multiple regression procedures. The response phenotypes, computed as the difference (Delta) between the posttraining and baseline measures, were adjusted for age, sex, and the baseline value. RESULTS: All submaximal exercise phenotypes measured at baseline and in response to training were characterized by a significant familial resemblance. Maximal heritabilities of the baseline phenotypes range from 48% to 74% with significant spouse, sibling, and parent-offspring correlations. The hypothesis of maternal inheritance where mother-offspring and sibling correlations were forced to be equal was found to fit the data for VO(2)60%, VO(2)80% and PO80%. For the response phenotypes, the maximal heritabilities tended to be lower (23--57%) with a significant maternal inheritance for Delta VO(2)60%, Delta PO60%, and Delta PO80%. CONCLUSION: These results suggest that the submaximal working capacities of sedentary subjects and their responses to endurance training are influenced by familial/genetic factors with a significant contribution of maternal inheritance.

Adult↗

PubMed2001

A genome-wide scan for abdominal fat assessed by computed tomography in the Québec Family Study.

To identify chromosomal regions harboring genes influencing the propensity to store fat in the abdominal area, a genome-wide scan for abdominal fat was performed in the Quebec Family Study. Cross-sectional areas of the amount of abdominal total fat (ATF) and abdominal visceral fat (AVF) were assessed from a computed tomography scan taken at L4-L5 in 521 adult subjects. Abdominal subcutaneous fat (ASF) was obtained by computing the difference between ATF and AVF. The abdominal fat phenotypes were adjusted for age and sex effects as well as for total amount of body fat (kilogram of fat mass) measured by underwater weighing, and the adjusted phenotypes were used in linkage analyses. A total of 293 microsatellite markers spanning the 22 autosomal chromosomes were typed. The average intermarker distance was 11.9 cM. A maximum of 271 sib-pairs were available for single-point (SIBPAL) and 156 families for multipoint variance components (SEGPATH) linkage analyses. The strongest evidence of linkage was found on chromosome 12q24.3 between marker D12S2078 and ASF (logarithm of odds [LOD] = 2.88). Another marker (D12S1045) located within 2 cM of D12S2078 also provided evidence of sib-pair linkage with ASF (P = 0.019), ATF (P = 0.015), and AVF (P = 0.0007). Other regions with highly suggestive evidence (P < 0.0023 or LOD > or =1.75) of multipoint linkage and evidence (P < 0.05) of single-point linkage, all for ASF, included chromosomes 1p11.2, 4q32.1, 9q22.1, 12q22-q23, and 17q21.1. Three of these loci (1p11.2, 9q22.1, and 17q21.1) are close to genes involved in the regulation of sex steroid levels, whereas two others (4q32.1 and 17q21.1) are in the proximity of genes involved in the regulation of food intake. This first genome-wide scan for abdominal fat assessed by computed tomography indicates that there may be several loci determining the propensity to store fat in the abdominal depot and that some of these loci may influence the development of diabetes in obese subjects.

Adipose Tissue↗

PubMed2000

A genome scan for renal function among hypertensives: the HyperGEN study.

Decreased renal function is often a complication of hypertension. Although it has been suggested that the response of the kidney to hypertension has an underlying genetic component, there is limited information suggesting that specific genetic regions or candidate genes contribute to the variability in creatinine clearance, a commonly used measure of kidney function. As part of the Hypertension Genetic Epidemiology Network (HyperGEN) study, creatinine clearance measurements were assessed in a large biracial sample of hypertensive siblings (466 African American subjects and 634 white subjects in 215 and 265 sibships, respectively). All participants were hypertensive before the age of 60 years, and the mean age of the siblings was 52 years among the African American subjects and 61 years among the white subjects. Two residual models were created for creatinine clearance: a minimally adjusted model (which included age and age(2)) and a fully adjusted model (which included age, age(2), lean body mass, pulse rate, pulse pressure, hormone-replacement therapy, educational status, and physical activity). Standardized residuals were calculated separately for men and women in both racial groups. The heritability of the residual creatinine clearance was 17% and 18% among the African American and white subjects, respectively. We conducted multipoint variance components linkage analysis using GENEHUNTER2 and 387 anonymous markers (Cooperative Human Linkage Center screening set 8). The best evidence for linkage in African American subjects was found on chromosome 3 (LOD = 3.61 at 214.6 cM, 3q27) with the fully adjusted model, and the best evidence in white subjects was found on chromosome 3 (LOD = 3.36 at 115.1 cM) with the minimally adjusted model. Positional candidate genes that are contained in and around the region on chromosome 3 (214.6 cM) that may contribute to renal function include enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase (EHHADH) and apolipoprotein D (ApoD). These findings suggest there may be genetic regions related to the variability of creatinine clearance among hypertensive individuals.

Black or African American↗

PubMed2000

Hostility, social support, and carotid artery atherosclerosis in the National Heart, Lung, and Blood Institute Family Heart Study.

This cross-sectional study investigates the association of hostility and social support (measured by standardized instruments) to carotid artery atherosclerosis in men and women with a high familial risk for coronary heart disease (CHD) and those with low to medium risk. The hypothesis was that high hostility and low social support would have a stronger association in subjects with a familial predisposition to CHD. There were 535 low- to medium-risk women, 491 low- to medium-risk men, 1,950 high-risk women, and 1,667 high-risk men in the study. The extent of carotid artery atherosclerosis was assessed by B-mode ultrasound imaging. A lesion was defined as an intimal-medial far wall thickness of 1 mm in the common, internal, or carotid bifurcation, or identification of plaque at any site. Odds ratios and their 95% confidence intervals were calculated using generalized estimating equations (GEE) for logistic regression. Family was specified as the clustering variable, and robust SEEs were obtained that account for dependence of the data within families. After controlling for age, education, body mass index, ever having smoked, ever drinking > 5 drinks a day, and metabolic index, hostility was significantly associated with increased odds of carotid lesions in only high-risk women. High-risk women showed a significantly reduced odds of carotid lesions with high social support, but the extent of this protection was reduced when age and education were included in the equation. A combination of high hostility and low social support was associated with higher odds than hostility alone in both high-risk men and women. These results suggest that women with a high familial predisposition for CHD may be more vulnerable to cardiovascular influences from hostility and social support than high-risk men or men and women with low to medium risk.

Adult↗

PubMed2000

Genome-wide linkage analysis of systolic and diastolic blood pressure: the Québec Family Study.

BACKGROUND: Blood pressure (BP), an important risk factor for coronary heart disease, is a complex trait with multiple genetic etiologies. While some loci affecting BP variation are known (eg, angiotensinogen), there are likely to be novel signals that can be detected with a genome scan approach. METHODS AND RESULTS: A genome-wide scan was performed in 125 random and 81 obese families participating in the Québec Family Study. A multipoint variance-components linkage analysis of 420 markers (353 microsatellites and 67 restriction fragment length polymorphisms) revealed several signals (P:<0.0023) for systolic BP on 1p (D1S551, ATP1A1), 2p (D2S1790, D2S2972), 5p (D5S1986), 7q (D7S530), 8q (CRH), and 19p (D19S247). Suggestive evidence (0.0023<P:<0.01) was found on 3q, 10p, 12p, 14q, and 22q. The results were encouraging for HSD3B1 (P:<0.03), AGT (P:<0.03), ACE (P:<0.02), and adipsin (P:<0.005) but null with regard to other candidates (eg, renin, and glucocorticoid and adrenergic receptors). CONCLUSIONS: Multiple linkage regions support the notion that risk for hypertension is due to multiple (ie, oligogenic) susceptibility loci. Comparisons across the complete, random, and obese samples suggest that some regions are specific to BP and others may involve obesity (eg, pleiotropy, epistasis, or gene-environment interaction). Some of these areas harbor known candidates. Others involve novel regions, some of which replicate previous reports and provide a focus for future studies to identify novel genes that influence interindividual variation in BP.

Age Distribution↗

PubMed2000

Family risk score of coronary heart disease (CHD) as a predictor of CHD: the Atherosclerosis Risk in Communities (ARIC) study and the NHLBI family heart study.

Family history of coronary heart disease (CHD) has been found to be a risk factor for CHD in numerous studies. Few studies have addressed whether a quantitative measure of family history of CHD (family risk score, FRS) predicts CHD in African Americans. This study assessed the association between FRS and incident CHD of participants, and the variation of the association by gender and race. Participants in the study were a biracial population-based cohort with 3,958 African Americans and 10,580 Whites aged 45-64 years old in the ARIC baseline survey (1987-1989). They were randomly selected from four U. S. communities. During follow-up (1987-1993), 352 participants experienced the onset of CHD. Incidence density of CHD (per 1,000 person-years) was 7.8 and 3.6 among African-American men (AAM) and women (AAW), and 7.2 and 2.2 among White men (WM) and women (WW). The hazard rate ratio (HRR) of CHD associated with one standard deviation increase of FRS was 1.52 in AAW, 1.46 in AAM, 1.41 in WW, and 1.68 in WM. The HRRs decreased 4.6% in AAW, 1.4% in WW, 5.7% in AAM, and 3.0% in WM, but increased 2.1% in AAM after adjustment for selected covariates. FRS predicts incident CHD in African Americans and Whites, men and women. The relation of FRS to incident CHD can be only partially explained by the selected risk factors in the biological causal pathways: IMT, T-G, LDL, HDL, Lp(a), fibrinogen and hypertension. No significant difference by race has been found in this study.

Age Factors↗

PubMed2000

Evidence for major genes influencing pulmonary function in the NHLBI family heart study.

Segregation analysis was performed on the pulmonary measures forced expiratory volume in one second (FEV1), forced vital capacity (FVC), and the ratio of FEV1/FVC in 455 randomly ascertained families from the NHLBI Family Heart Study (FHS). Gender specific standardized residuals were used as the phenotypic variable in both familial correlation and segregation analyses. These residuals represented adjustments for the effects of age, age(2), age(3), Body Mass Index (BMI, kg/m(2)), height, the ratio of waist to hip measurements (WHR), the presence of coronary heart disease, smoking history, and pack years for current smokers. Sibling correlations were not different from parent-offspring correlations for all three traits, and heritability estimates for FEV1, FVC, and the FEV1/FVC ratio were 0. 515, 0.540, and 0.449, respectively. Segregation analysis of FEV1, a trait that measures airflow, indicated that a dominant major gene best fits the data, although a residual familial correlation supports the presence of an additional polygenic or common environmental component. For FVC, a trait that measures lung volume, alternative models could not be statistically differentiated, but the transmission probabilities do not support a Mendelian major gene. The best model for FEV1/FVC ratio is a non-Mendelian codominant model, perhaps due to the mixing of the individual underlying distributions influencing airflow and lung volume. These results support the hypothesis that complex relationships exist for lung function traits and that multiple genes and environmental factors influence lung function.

Anthropometry↗

PubMed2000

A single, sequential, genome-wide test to identify simultaneously all promising areas in a linkage scan.

Inflation of type I error occurs when conducting a large number of statistical tests in genome-wide linkage scans. Stringent alpha-levels protect against the high numbers of expected false positives but at the cost of more false negatives. A more balanced tradeoff is provided by the theory of sequential analysis, which can be used in a genome scan even when the data are collected using a fixed-sample design. Sequential tests allow complete, simultaneous control of both the type I and II errors of each individual test while using the smallest possible sample size for analysis. For fixed samples, the excess N "saved" can be used in a confirmatory, replication phase of the original findings. Using the theory of sequential multiple decision procedures [Bechhoffer et al., 1968], we can replace the series of individual marker tests with a new single, simultaneous genome-wide test that has multiple possible outcomes and partitions all markers into two subsets: the "signal" versus the "noise," with an a priori specifiable genome-wide error rate. These tests are demonstrated for the Haseman-Elston approach, are applied to real data, and are contrasted with traditional fixed-sampling tests in Monte Carlo simulations of repeated genome-wide scans. The method allows efficient identification of the true signals in a genome scan, uses the smallest possible sample sizes, saves the excess to confirm those findings, controls both types of error, and provides one elegant solution to the debate over the best way to balance between false positives and negatives in genome scans.

Dopamine beta-Hydroxylase↗

PubMed2000

Familial history of metabolic disorders and the multiple metabolic syndrome: the NHLBI family heart study.

UNLABELLED: A case-control study was conducted to investigate the association between family history of obesity, hypertension, and diabetes and the co-occurrence of metabolic disorders associated with the multiple metabolic syndrome (MMS). Included were 1,448 African and European American men and women aged 48-71 who participated in both the third cohort examination of the Atherosclerosis Risk in Communities study, 1992-1994, and phase I of the Family Heart Study 1993-1995. The joint occurrence of hypertension, dyslipidemia, and diabetes or impaired fasting glucose in an individual determined his/her status of "affected" (MMS: n = 97), while the absence of these three metabolic disorders determined his/her status of "unaffected" ( CONTROL: n = 527). First-degree relatives provided the information to calculate family risk scores (FRSs) for the phenotypes under study: obesity, diabetes and hypertension. Although the majority of cases were obese (76.3%), family history of obesity was associated only weakly with the MMS, while family history of diabetes, or hypertension was associated significantly with the MMS (controlling for age, race, gender, and sampling group). Obesity of cases and controls modified the strength of these associations-odds ratios were 2.5(95% CI:1.1-6.1) and 2.9(95% CI:1.2-7.0) for the diabetes and hypertension FRSs in the non-obese, while in obese individuals the respective odds ratios were 1.6(95% CI:0.9-2.8) and 1.7(95% CI:0.9-3.1). These results may imply that obesity, whether familial or environmental in nature, is associated with the development of the MMS, while in non-obese individuals a family history of diabetes, hypertension, or obesity is a marker of genetic predisposition to components of the MMS.

Aged↗

PubMed2000

Smoking influences the association between apolipoprotein E and lipids: the National Heart, Lung, and Blood Institute Family Heart Study.

Apolipoprotein E allele 4 (apo epsilon4) and smoking each have been associated with an unfavorable lipid profile. We used data collected on 1,472 subjects in the National Heart, Lung, and Blood Institute Family Heart Study to assess whether smoking interacts with apo epsilon4 to influence the levels of plasma lipids. We dichotomized smoking and apo epsilon4 and used analysis of covariance to estimate the means of lipids. Smokers had lower body mass index, were younger, and consumed less fruits and vegetables. Among individuals without apo epsilon4, comparing nonsmokers with smokers, mean low density lipoprotein cholesterol (LDL) was 129.3 and 134.4 mg/dL, respectively, for women and 126.1 and 127.6 mg/dL, respectively, for men. Among subjects with an apo epsilon4 allele, corresponding means were 132.0, and 152.9 mg/dL, respectively, for women and 131.3 and 137.3 mg/dL, respectively, for men (Pfor interaction <0.001 for women and 0.11 for men). A similar interaction was observed for total cholesterol among women (P = 0.02). This study shows a statistically significant effect modification of the relation of apo epsilon4 to LDL and total cholesterol by smoking among women. Smoking may enhance genetic susceptibility to an unfavorable lipid profile among subjects with apo epsilon4.

Adult↗