Logistic model estimation of death attributable to risk factors for cardiovascular disease in Evans County, Georgia

Associations of broad eating disorder symptoms with later alcohol problems in Finnish adult twins: A nationwide 10-year follow-up

OBJECTIVE

Eating disorders are associated with subsequent alcohol problems, but it is not known whether this association also extends to broader eating disorder symptoms not captured by clinical diagnoses. We assessed the longitudinal association of broad eating disorder symptoms with alcohol problems in a nationwide twin sample (FinnTwin16).

METHODS

Finnish women (N = 1905) and men (N = 1449) self-reported their eating disorder symptoms using the Eating Disorder Inventory-2 Bulimia, Drive for Thinness, and Body Dissatisfaction subscales at the mean age of 24.4 years in 2000-2003. A subsample of participants also completed items on drive for muscularity, height dissatisfaction, and muscle-enhancing supplement use. Alcohol problems were assessed 10 years later at the age of 34.1 in 2010-2012 with the Rutgers Alcohol Problem Index.

RESULTS

Eating disorder symptoms were associated with later alcohol problems (odds ratio per point increase 1.02-1.18). Bulimia showed stronger associations among men than women (p for interaction .012). Drive for muscularity and height dissatisfaction were also associated with later alcohol problems, but supplement use was not. When accounting for baseline alcohol problems, only Bulimia (among women and men) and Drive for Thinness (among men) were significantly associated with later alcohol problems. Bulimia was also significantly associated with later alcohol problems in within-twin-pair analyses among dizygotic twins, but not among monozygotic twins.

DISCUSSION

In a longitudinal setting, eating disorder symptoms were associated with later alcohol problems. Bulimic symptoms were a stronger risk factor for men than women. These associations may be attributable to baseline alcohol problems, childhood environment and genetic liability.

PUBLIC SIGNIFICANCE

This study found that both young adult women and men with broad eating disorder symptoms are at a higher risk of alcohol-related problems than those without such symptoms. Men with bulimic symptoms were at a particularly high risk. These findings emphasize the need for better prevention and treatment of disordered eating, body image concerns and alcohol problems for both young adult women and men.

A Bayesian approach for estimating the partial potential impact fraction with exposure measurement error under a main study/internal validation design

The partial potential impact fraction describes the proportion of disease cases that can be prevented if the distribution of modifiable continuous exposures is shifted in a population, while other risk factors are not modified. It is a useful quantity for evaluating the burden of disease in epidemiologic and public health studies. When exposures are measured with error, the partial potential impact fraction estimates may be biased, which necessitates methods to correct for the exposure measurement error. Motivated by the health professionals follow-up study, we develop a Bayesian approach to adjust for exposure measurement error when estimating the partial potential impact fraction under the main study/internal validation study design. We adopt the reclassification approach that leverages the strength of the main study/internal validation study design and clarifies transportability assumptions for valid inference. We assess the finite-sample performance of both the point and credible interval estimators via extensive simulations and apply the proposed approach in the health professionals follow-up study to estimate the partial potential impact fraction for colorectal cancer incidence under interventions exploring shifting the distributions of red meat, alcohol, and/or folate intake.

Semiparametric estimation of the attributable fraction when there are interactions under monotonicity constraints

BACKGROUND

The population attributable fraction (PAF) is the fraction of disease cases in a sample that can be attributed to an exposure. Estimating the PAF often involves the estimation of the probability of having the disease given the exposure while adjusting for confounders. In many settings, the exposure can interact with confounders. Additionally, the exposure may have a monotone effect on the probability of having the disease, and this effect is not necessarily linear.

METHODS

We develop a semiparametric approach for estimating the probability of having the disease and, consequently, for estimating the PAF, controlling for the interaction between the exposure and a confounder. We use a tensor product of univariate B-splines to model the interaction under the monotonicity constraint. The model fitting procedure is formulated as a quadratic programming problem, and, thus, can be easily solved using standard optimization packages. We conduct simulations to compare the performance of the developed approach with the conventional B-splines approach without the monotonicity constraint, and with the logistic regression approach. To illustrate our method, we estimate the PAF of hopelessness and depression for suicidal ideation among elderly depressed patients.

RESULTS

The proposed estimator exhibited better performance than the other two approaches in the simulation settings we tried. The estimated PAF attributable to hopelessness is 67.99% with 95% confidence interval: 42.10% to 97.42%, and is 22.36% with 95% confidence interval: 12.77% to 56.49% due to depression.

CONCLUSIONS

The developed approach is easy to implement and supports flexible modeling of possible non-linear relationships between a disease and an exposure of interest.

Attributable fraction for multiple risk factors: Methods, interpretations, and examples

The attributable fraction is the candidate tool to quantify individual shares of each risk factor on the disease burden in a population, expressing the proportion of cases ascribable to the risk factors. The original formula ignored the presence of other factors (i.e. multiple risk factors and/or confounders), and several adjusting methods for potential confounders have been proposed. However, crude and adjusted attributable fractions do not sum up to their joint attributable fraction (i.e. the number of cases attributable to all risk factors together) and their sum may exceed one. A different approach consists of partitioning the joint attributable fraction into exposure-specific shares leading to sequential and average attributable fractions. We provide an example using Italian case-control data on oral cavity cancer comparing crude, adjusted, sequential, and average attributable fractions for smoking and alcohol and provide an overview of the available software routines for their estimation. For each method, we give interpretation and discuss shortcomings. Crude and adjusted attributable fractions added up over than one, whereas sequential and average methods added up to the joint attributable fraction = 0.8112 (average attributable fractions for smoking and alcohol were 0.4894 and 0.3218, respectively). The attributable fraction is a well-known epidemiological measure that translates risk factors prevalence and disease occurrence in useful figures for a public health perspective. This work endorses their proper use and interpretation.

Estimation of causal effect measures with the R-package stdReg

Measures of causal effects play a central role in epidemiology. A wide range of measures exist, which are designed to give relevant answers to substantive epidemiological research questions. However, due to mathematical convenience and software limitations most studies only report odds ratios for binary outcomes and hazard ratios for time-to-event outcomes. In this paper we show how logistic regression models and Cox proportional hazards regression models can be used to estimate a wide range of causal effect measures, with the R-package stdReg. For illustration we focus on the attributable fraction, the number needed to treat and the relative excess risk due to interaction. We use two publicly available data sets, so that the reader can easily replicate and elaborate on the analyses. The first dataset includes information on 487 births among 188 women, and the second dataset includes information on 2982 women diagnosed with primary breast cancer.

Spatial variation in attributable risks

The attributable risk (AR) measures the contribution of a particular risk factor to a disease, and allows estimation of disease rates specific to that risk. While previous studies consider variability in ARs over demographic categories, this paper considers the extent of spatial variability in ARs estimated from multilevel data with confounders both at individual and geographic levels. A case study considers the AR for diabetes in relation to elevated BMI, and area rates for diabetes attributable to excess weight. Contextual adjustment includes known area variables, and unobserved spatially clustered influences, while spatial heterogeneity (effect modification) is considered in terms of varying effects of elevated BMI by neighbourhood deprivation category. The application is to patient register data in London, with clear evidence of spatial variation in ARs, and in small area diabetes rates attributable to excess weight.

From exposures to population interventions: pregnancy and response to HIV therapy

Many epidemiologic studies identify contrasts between an "always-exposed" population and a "never-exposed" population. Such "exposure effects" are perhaps most valuable in discussing individual lifestyle changes, or in clinical care; they may be less valuable in estimating the potential effects of realistic public health interventions. Various methods, among them population attributable fractions and generalized impact fractions, attempt to obtain more policy-relevant estimates of "population intervention" effects, but such methods remain rare in the epidemiologic literature. Here, we describe the use of the parametric g-formula as a tool for the estimation of population intervention effects in longitudinal data. Our discussion is motivated by a previous study of the effect of incident pregnancy on time to virological failure among human immunodeficiency virus-positive women initiating antiretroviral therapy in South Africa between 2004 and 2011. We show that 1) interventional estimates of effect can be estimated in longitudinal data using the parametric g-formula and 2) exposure effects and population interventional effects can have dramatically different interpretations and magnitudes in real-world data. Epidemiologists should consider estimating interventional effects in addition to exposure effects; doing so would allow the results of epidemiologic studies to be more immediately relevant to policy-makers and to implementation science efforts.

Association between human papillomavirus vaccine uptake and cervical cancer screening in the Netherlands: implications for future impact on prevention

Several countries recently added human papillomavirus (HPV) vaccination to cervical cancer screening in the effort to prevent cervical cancer. They include the Netherlands, where both programs are free. To estimate their combined future impact on cancer prevention, information is needed on the association between participation in vaccination now and in screening in the future and on what groups are at risk for nonparticipation. We studied the association between participation in screening by mothers and in vaccination by their daughters. Girls' vaccination status was matched by house-address with their mothers' screening participation. We estimated the effect on cancer incidence by means of computer simulation. We investigated risk groups for nonparticipation using multivariable multilevel logistic regression and calculated population-attributable fractions. Our results, based on 89% of girls invited for vaccination in 2009 (n = 337,368), show that vaccination status was significantly associated with mothers' screening participation (odds ratio: 1.54 [95% confidence interval: 1.51-1.57]). If a mother's screening is taken as proxy of a girl's future screening, only 13% of the girls will not participate in either program compared to 23% if screening alone is available. The positive association between vaccination and screening resulted in slightly lower model estimates of the impact of vaccination on cancer incidence, compared to estimates assuming no association. Girls with nonwestern ethnicities, with young mothers, who live in urban areas with low socioeconomic status, are at risk for nonparticipation. A significant part of potential nonscreeners may be reached through HPV vaccination. Estimates made before vaccination was introduced only slightly overestimated its impact on cervical cancer incidence.

Efficient estimation of the attributable fraction when there are monotonicity constraints and interactions

The PAF for an exposure is the fraction of disease cases in a population that can be attributed to that exposure. One method of estimating the PAF involves estimating the probability of having the disease given the exposure and confounding variables. In many settings, the exposure will interact with the confounders and the confounders will interact with each other. Also, in many settings, the probability of having the disease is thought, based on subject matter knowledge, to be a monotone increasing function of the exposure and possibly of some of the confounders. We develop an efficient approach for estimating logistic regression models with interactions and monotonicity constraints, and apply this approach to estimating the population attributable fraction (PAF). Our approach produces substantially more accurate estimates of the PAF in some settings than the usual approach which uses logistic regression without monotonicity constraints.

Participation in and attitude towards the national immunization program in the Netherlands: data from population-based questionnaires

BACKGROUND

Knowledge about the determinants of participation and attitude towards the National Immunisation Program (NIP) may be helpful in tailoring information campaigns for this program. Our aim was to determine which factors were associated with nonparticipation in the NIP and which ones were associated with parents' intention to accept remaining vaccinations. Further, we analyzed possible changes in opinion on vaccination over a 10 year period.

METHODS

We used questionnaire data from two independent, population-based, cross-sectional surveys performed in 1995-96 and 2006-07. For the 2006-07 survey, logistic regression modelling was used to evaluate what factors were associated with nonparticipation and with parents' intention to accept remaining vaccinations. We used multivariate multinomial logistic regression modelling to compare the results between the two surveys.

RESULTS

Ninety-five percent of parents reported that they or their child (had) participated in the NIP. Similarly, 95% reported they intended to accept remaining vaccinations. Ethnicity, religion, income, educational level and anthroposophic beliefs were important determinants of nonparticipation in the NIP. Parental concerns that played a role in whether or not they would accept remaining vaccinations included safety of vaccinations, maximum number of injections, whether vaccinations protect the health of one's child and whether vaccinating healthy children is necessary. Although about 90% reported their opinion towards vaccination had not changed, a larger proportion of participants reported to be less inclined to accept vaccination in 2006-07 than in 1995-96.

CONCLUSION

Most participants had a positive attitude towards vaccination, although some had doubts. Groups with a lower income or educational level or of non-Western descent participated less in the NIP than those with a high income or educational level or indigenous Dutch and have been less well identified previously. Particular attention ought to be given to these groups as they contribute in large measure to the rate of nonparticipation in the NIP, i.e., to a greater extent than well-known vaccine refusers such as specific religious groups and anthroposophics. Our finding that the proportion of the population inclined to accept vaccinations is smaller than it was 10 years ago highlights the need to increase knowledge about attitudes and beliefs regarding the NIP.

Prevalence of hepatitis B virus infection in The Netherlands in 1996 and 2007

We aimed to assess differences in the prevalence of hepatitis B virus (HBV) infection in The Netherlands between 1996 and 2007, and to identify risk factors for HBV infection in 2007. Representative samples of the Dutch population in 1996 and 2007 were tested for antibodies to hepatitis B core antigen (anti-HBc), hepatitis B surface antigen (HBsAg) and HBV-DNA. In 2007, the weighted anti-HBc prevalence was 3·5% (95% CI 2·2-5·5) and the HBsAg prevalence was 0·2% (95% CI 0·1-0·4). In indigenous Dutch participants, the anti-HBc prevalence was lower in 2007 than in 1996 (P=0·06). First-generation migrants (FGMs) had a 13-fold greater risk of being HBsAg- and/or HBV-DNA-positive than indigenous Dutch participants. In indigenous Dutch participants, risk factors for anti-HBc positivity were older age and having received a blood product before 1990. In FGMs, being of Asian origin was a risk factor. In second-generation migrants, having a foreign-born partner and injecting drug use were risk factors. FGMs are the main target group for secondary HBV prevention in The Netherlands.

Biostatistics and epidemiology: measuring the risk attributable to an environmental or genetic factor

Disease frequency is measured through estimating incidence rates or disease risk. Several measures are used for assessing exposure-disease association, with adjusted estimates based on standardization, stratification, or more flexible regression techniques. Several measures are available to assess an exposure impact in terms of disease occurrence at the population level, including the commonly used attributable risk (AR). Adjusted AR estimation relies on stratification or regression techniques. Sequential and partial ARs have been proposed to handle the situation of multiple exposures and circumvent the associated non-additivity problem. Despite remaining issues in properly interpreting AR, AR remains a useful guide to assess prevention strategies.

Impact of asthma on self-reported health status and quality of life: a population based study of Australians aged 18-64

BACKGROUND

The impact of asthma has traditionally been measured in terms of the prevalence of the disease, mortality rates, and levels of healthcare utilisation, particularly hospital admissions. However, the impact of asthma extends beyond these outcomes to include effects on lifestyle, well being, and perceived health status.

METHODS

Information on self-reported current asthma status, arthritis and diabetes as well as measures of life satisfaction, self-assessed health status, psychological distress, and interference with usual activities was obtained for 14 641 respondents aged 18-64 years in the 2001 National Health Survey of the general population in all states and territories in Australia. Log linear models were fitted separately for each of the dichotomised quality of life measures as dependent variables. The estimates of the adjusted rate ratio obtained from each model were used to compute the population attributable fraction (PAF) of self-reported asthma, arthritis, and diabetes for each of the health status and quality of life measures.

RESULTS

The presence of asthma accounted for 3.18% (95% CI 2.13 to 4.23) of people reporting poor life satisfaction, 8.12% (95% CI 6.57 to 9.67) of people reporting poor health status, 5.90% (95% CI 4.19 to 7.61) of people reporting high psychological distress, and 3.58% (95% CI 2.16 to 5.01) of people reporting any reduced activity days. The proportions of people with these adverse health states attributable to asthma were higher than the proportions attributable to diabetes but lower than the proportions attributable to arthritis.

CONCLUSION

Asthma is an important contributor to the burden of ill health and impaired quality of life in the community. A strategic approach is needed to develop and implement strategies to address the impact of asthma on quality of life.

Epidemiologic appraisal of hypertension as a coronary risk factor in the elderly

Five decades of epidemiologic research has established elevated blood pressure as a major contributor to atherosclerotic cardiovascular diseases in the elderly, including coronary heart disease. Clinicians formerly favored the diagnosis and treatment of hypertension in terms of the diastolic blood pressure and categorical "hypertension." Epidemiologic data now emphasize the essential role of systolic blood pressure, pulse pressure, and a graded influence of blood pressure, even within the high-normal range. The risk of coronary heart disease, the most common lethal sequela of hypertension, increases with the extent of risk factor clustering. Among hypertensive persons, about 39% of coronary events in men and 68% in women are attributable to the presence of two or more additional risk factors. When risk factor clustering is associated with glucose intolerance, obesity, and dyslipidemia, it may be attributed to insulin resistance promoted by abdominal obesity. Other hazardous influences often accompanying hypertension in the elderly are the presence of an elevated heart rate, elevated levels of fibrinogen, and left ventricular hypertrophy. Because clustering with other risk factors is characteristic of hypertension in the elderly, it is essential to screen for them and for the presence of comorbid cardiovascular diseases, target organ disease, and subclinical vascular disease likely to be present. Multivariate risk assessment profiles enable global estimation of hypertensive risk of developing coronary heart disease. Hypertensive elderly patients are more appropriately targeted for antihypertensive therapy by such risk stratification than by relying solely on the severity of the blood pressure elevation. The goal of therapy should be to improve the multivariate risk profile as well as the level of the blood pressure.

Attributable fractions: bias from broad definition of exposure

In certain special situations, simplification of an exposure measure into a dichotomy results in no bias from nondifferential misclassification when estimating the attributable fraction for "any exposure." This fact has led to recommendations to use a broad definition of exposure when estimating attributable fractions. I here review the assumptions underlying exposure simplification, focusing on the assumptions that the source and target populations have the same exposure distribution and that complete risk removal is possible. I argue that attributable fraction estimates based on dichotomization can be especially sensitive to violations of these assumptions, and hence misleading for projecting the impact of exposure reduction. I conclude that it is important to obtain and use detailed exposure and covariate information for attributable-fraction estimation.

A review of adjusted estimators of attributable risk

This paper reviews adjusted methods of estimation of attributable risk (AR), that is methods that allow one to obtain estimates of AR while controlling for other factors. Estimability and basic principles of AR estimation are first considered and the rationale for adjusted AR estimators is discussed. Then, adjusted AR estimators are reviewed focusing on cross-sectional, cohort and case-control studies. Two inconsistent adjusted estimators are briefly commented upon. Next, adjusted estimators based on stratification, namely the weighted-sum and Mantel-Haenszel (MH) approaches, are reviewed and contrasted. It appears that the weighted-sum approach, which allows for full interaction between exposure and adjustment factors, can be affected by small-sample bias. By contrast, the MH approach, which rests on the assumption of no interaction between exposure and adjustment factors may be misleading if interaction between exposure and adjustment factors is present. Model-based adjusted estimators represent a more general and flexible approach that includes both stratification approaches as special cases and offers intermediate options. Bruzzi et al.'s and Greenland and Drescher's estimators are reviewed and contrasted. Finally, special problems of adjusted estimation are considered, namely estimation from case-cohort data, estimation for risk factors with multiple levels, for multiple risk factors, for recurrent events, estimation of the prevented and preventable fractions, and estimation of the generalized impact fraction. Comments on future directions are presented.

Estimation of population attributable fractions from fitted incidence ratios and exposure survey data, with an application to electromagnetic fields and childhood leukemia

Standard presentations of epidemiological results focus on incidence-ratio estimates derived from regression models fit to specialized study data. These data are often highly nonrepresentative of populations for which public-health impacts must be evaluated. Basic methods are provided for interval estimation of attributable fractions from model-based incidence-ratio estimates combined with independent survey estimates of the exposure distribution in the target population of interest. These methods are illustrated in estimation of the potential impact of magnetic-field exposures on childhood leukemia in the United States, based on pooled data from 11 case-control studies and a U.S. sample survey of magnetic-field exposures.

Physical load in daily life and low back problems in the general population-The MORGEN study

BACKGROUND

We studied the contribution of physical load in daily activities, including activities in work, housekeeping, and leisure time, to the burden of low back problems (LBP) in the population.

METHODS

Logistic regression models were used to calculate the association between physical load and several LBP parameters as assessed by questionnaire in a cross-sectional study on 22,415 randomly selected men and women in The Netherlands, controlling for well-known LBP determinants. The population attributable risk (PAR) percentage was estimated with the elimination method using the logistic model.

RESULTS

Half of the population reported LBP during the past year and 19% chronic LBP. Activities characterized by an awkward posture, by the same posture for a long time, or by often bending and rotating the trunk increased the risk for LBP, with ORs between 1.1 and 1.6. More than 13% of the 1-year prevalence of LBP could be contributed to these activities. This PAR was higher for those belonging to the working population, for women, and for the more severe LBP parameters.

CONCLUSION

Because LBP present such a large public health problem, the estimated potential impact of eliminating (the unhealthy effect of) physical load is substantial. To assess the real health gain, more insight is necessary into the causality of the relationship and into effective preventive measures.

Bayesian inference for a generalized population attributable fraction: the impact of early vitamin A levels on chronic lung disease in very low birthweight infants

In this paper, the population attributable fraction is studied using the potential responses framework of Rubin's causal model. This framework facilitates definition of a general measure of population attributable effect which can accommodate many-valued and multivariate exposures as well as many-valued responses. Inferential issues are considered from the Bayesian perspective. Finite population inference is emphasized with inference in the case of a fully observed population given particular attention. The key inferential issue concerns computation of the posterior distribution of unobserved potential responses, given observed responses, exposures and covariates. A dependency on model parameters about which observed data are uninformative is highlighted and this reflects the unobservable nature of causal effects. In an application to a small cohort study of respiratory problems in very low birthweight infants, posterior inferences were found to be insensitive to assumptions concerning the joint distribution of potential response variables but sensitive to the assumption of weak ignorability, a weaker form of the more familiar assumption of no confounding by omitted covariates. In a model-based set-up, the weak ignorability assumption is identified with setting a model parameter to zero, and consequently uncertainty concerning this assumption can, in principle, be handled via the prior distribution for the model parameters.

Elevated systolic blood pressure as a cardiovascular risk factor

This paper reviews the evolution of attitudes toward the treatment and diagnosis of hypertension. In particular, there is a growing realization that elevated systolic pressure may be a more valuable measurement in evaluating and controlling hypertension than is generally acknowledged. A large number of epidemiologic studies in a wide variety of populations have revealed that systolic blood pressure exerts a stronger influence than diastolic blood pressure. The largest of these, the Framingham Heart Study, showed that in subjects with systolic hypertension, diastolic blood pressure was only weakly related to the risk of cardiovascular events, but in those with diastolic hypertension, the risk of these events was strongly influenced by the level of systolic pressure. Furthermore, cardiovascular event rates were found to increase steeply with systolic pressure and were higher in cases of isolated systolic hypertension than diastolic hypertension. Clinical trials produced similar results, again suggesting that a greater reliance should be placed on systolic pressure in evaluating the risk of cardiovascular problems. This review concludes that the health community needs to be reeducated to consider the importance of systolic and diastolic blood pressure in assessing appropriate management strategies for hypertensive patients.

A review on ethnic differences in plasma triglycerides and high-density-lipoprotein cholesterol: is the lipid pattern the key factor for the low coronary heart disease rate in people of African origin?

Black people in the UK, in the Caribbean, and to a lesser extent in the USA, experience coronary heart disease events at different rates than white people. Despite having higher prevalence of hypertension, cigarette smoking and diabetes, black males have significantly lower coronary heart disease rates than white males, whereas no significant differences have been detected in females. The only known risk factor differences that could account for the difference in CHD rates are higher HDL cholesterol and lower triglycerides that are seen in blacks compared with whites. Obesity and, in particular abdominal obesity, seems to determine TG and HDL cholesterol levels: black males are less centrally obese than whites, while total adiposity and central distribution of fat is more predominant in black females compared with white females. We propose that the less degree of abdominal adiposity observed in black males is related with an increased anti-lipolytic effect of insulin, which could account for low triglycerides and high HDL cholesterol levels, and consequently explain the higher protection from coronary heart disease experienced by black males compared with whites and black females.

The prevalence of hypertension in seven populations of west African origin

OBJECTIVES

This study was undertaken to describe the distribution of blood pressures, hypertension prevalence, and associated risk factors among seven populations of West African origin.

METHODS

The rates of hypertension in West Africa (Nigeria and Cameroon), the Caribbean (Jamaica, St. Lucia, Barbados), and the United States (metropolitan Chicago, Illinois) were compared on the basis of a highly standardized collaborative protocol. After researchers were given central training in survey methods, population-based samples of 800 to 2500 adults over the age of 25 were examined in seven sites, yielding a total sample of 10014.

RESULTS

A consistent gradient of hypertension prevalence was observed, rising from 16% in West Africa to 26% in the Caribbean and 33% in the United States. Mean blood pressures were similar among persons aged 25 to 34, while the increase in hypertension prevalence with age was twice as steep in the United States as in Africa. Environmental factors, most notably obesity and the intake of sodium and potassium, varied consistently with disease prevalence across regions.

CONCLUSION

The findings demonstrate the determining role of social conditions in the evolution of hypertension risk in these populations.

Outcomes of black and white hypertensive individuals after 30 years of follow-up

Outcomes of a 30-year follow-up for the participants of the Charleston Heart Study were studied with elevated blood pressure assessed using various classifications. The traditional categories of > or = 140/90 mmHg, > or = 160/95 mmHg and isolated systolic hypertension, as well as high normal and the four stages of high blood pressure were utilized in analyses. Prevalence rates of hypertension were, in general, higher among blacks compared to whites. Blacks had higher prevalence rates of hypertension and greater prevalence of high blood pressure at younger ages. Risk ratios were higher for black and white hypertensives than their normotensive counterparts. Blacks were found to have substantially higher population attributable risk proportions, particularly at the higher blood pressure categories. The results suggest that the standard clinical classifications of hypertension as related to mortality are appropriate for blacks and white.

Avoidable hospitalizations and socio-economic status in Galveston County, Texas

Population-based hospitalization rates for preventable conditions are derived for a low-income population in Galveston County, Texas using discharge data from four area hospitals and demographic data from the U.S. Census. Comparisons are made with similar rates for the general populations of two eastern states (Maryland and Massachusetts) and New York City. Results support the hypothesis that low-income persons lack access to primary care, leading to higher rates of hospitalization for preventable conditions. Alternative explanations for differences found are examined.

Echocardiographic correlates of left ventricular structure among 844 mildly hypertensive men and women in the Treatment of Mild Hypertension Study (TOMHS)

BACKGROUND

Echocardiography provides a noninvasive means of assessing left ventricular (LV) structure and evidence of LV wall remodeling in hypertensive persons. The relation of demographic, biological, and other factors with LV structure can be assessed.

METHODS AND RESULTS

LV structure was assessed by M-mode echocardiograms for 511 men and 333 women with mild hypertension (average blood pressure, 140/91 mm Hg). Measurements of LV wall thicknesses and internal dimensions were made, and estimates of LV mass indexes and other derivations of structure were calculated. LV hypertrophy criteria were based on previously reported echocardiographic population studies of normal subjects. These measures were compared by age, sex, race, body mass index, systolic blood pressure, antihypertensive drug use, physical activity, alcohol intake, cigarette smoking, and urinary sodium excretion. Despite virtual absence of ECG-determined LV hypertrophy, 13% of men and 20% of women had echocardiographically determined LV hypertrophy indexed by body surface area (g/m2), and 24% of men and 45% of women had LV hypertrophy indexed by height (g/m). Black participants had slightly higher mean levels of wall thickness than nonblack participants but similar LV mass. Systolic blood pressure and urinary sodium excretion were significantly and independently associated with LV mass index and LV hypertrophy using both g/m2 and g/m. Body mass index was significantly related to LV mass index and LV hypertrophy using g/m. Smoking was significantly associated with LV mass index, i.e., using continuous measurement but not using the dichotomy for LV hypertrophy.

CONCLUSIONS

This study of a large population of men and women with mild primary hypertension, largely without ECG evidence of LV hypertrophy, showed a substantial percentage of participants with echocardiographically determined LV hypertrophy. LV mass indexes correlated positively with systolic blood pressure, body mass index, urinary sodium excretion, and smoking.

Ischemic heart disease prevention: estimating the impact of interventions

The potential impact of ischemic heart disease intervention programs has usually been assessed using the dichotomy between those programs targeted to high risk groups and those that are population based, but this distinction does not adequately describe the spectrum of possibilities. Using data from the National Health and Nutrition Examination Survey Epidemiologic Follow-up Study (NHEFS), we assessed the effect of a spectrum of 27 potential interventions on mortality reduction and on an Intervention Index (defined as the number of persons whose risk must change to prevent one death). Using combinations of cholesterol reductions of 20% and decreases in the prevalence of smoking and hypertension of 50%, reductions in mortality varied from 1 to 27% and the Intervention Index varied from 26 to 520. A number of potential interventions were equivalent in their mortality reduction of their Intervention Indexes, despite their affecting differing proportions of the population. The Intervention Index provides some measure of the relative efficiency of programs and points to the comparability of different interventive approaches. In addition, this analysis suggests that the potential impact of intervention programs on mortality will be modest, but that a focus on certain subgroups, such as those aged 40-59 years, can achieve substantial results within those groups, even though the population effect would be minimal.

Systolic hypertension in the elderly

Historically diastolic blood pressure (BP) rather than systolic BP has been regarded clinically as the more important component related to subsequent hypertensive morbidity and mortality, and treatment has thus been directed towards lowering the diastolic BP. Observational studies across many different populations have related cerebrovascular disease and death more to the systolic BP, which appears selectively to increase as the population ages. Isolated systolic hypertension (ISH), therefore, may be more prevalent as westernized societies become older. Those affected with ISH suffer a two- to fivefold increase in rates of stroke and ischemic heart disease compared to normotensives. Currently no clinical trials data exist for ISH showing the efficacy of antihypertensive therapy upon final morbidity and mortality, but a large-scale multicenter clinical trial, the Systolic Hypertension in the Elderly Program (SHEP), is currently underway in the United States. Results are expected in the early 1990s. If the results of this trial confirm the efficacy of treating ISH, the therapeutic challenge of ISH will be to selectively decrease systolic BP without undue side effects.

The effect of gender on the probability of myocardial infarction among emergency department patients with acute chest pain: a report from the Multicenter Chest Pain Study Group

OBJECTIVE

To identify differences in the incidences of myocardial infarction in women and men with chest pain.

DESIGN

Prospective multicenter cohort study.

SETTING

Emergency rooms of three university and four community hospitals.

PATIENTS

7,734 emergency room patients with acute chest pain.

MEASUREMENTS AND MAIN RESULTS

Myocardial infarction was diagnosed in 10% of the 3,896 women, compared with 19% of the 3,838 men, yielding an age-adjusted relative risk of myocardial infarction for women of 0.54 (95% confidence interval 0.48, 0.60). Physicians were equally adept at admitting women and men with myocardial infarctions, but men without myocardial infarction or unstable angina were significantly more likely to be admitted than were women without these diagnoses. Most clinical and electrocardiographic features indicating a risk of myocardial infarction were present in both women and men, but several high-risk features were less commonly present in women. After adjusting for the other factors that correlate with each patient's probability of having acute myocardial infarction, the relative risk of myocardial infarction was the same in women as men when the emergency department electrocardiogram showed the classic changes associated with acute myocardial infarction, but the risk was 40% lower in women when such electrocardiographic changes were not present.

CONCLUSIONS

Clinical features that predict myocardial infarction in men predict myocardial infarction in women to a similar extent. However, female gender is associated with about a 40% lower rate of myocardial infarction except when classic electrocardiographic evidence is present on the emergency department electrocardiogram.

Sample size and power based on the population attributable fraction

Most methods for calculating sample size use the relative risk (RR) to indicate the strength of the association between exposure and disease. For measuring the public health importance of a possible association, the population attributable fraction (PAF)--the proportion of disease incidence in a population that is attributable to an exposure--is more appropriate. We determined sample size and power for detecting a specified PAF in both cohort and case-control studies and compared the results with those obtained using conventional estimates based on the relative risk. When an exposure is rare, a study that has little power to detect a small RR often has adequate power to detect a small PAF. On the other hand, for common exposures, even a relatively large study may have inadequate power to detect a small PAF. These comparisons emphasize the importance of selecting the most pertinent measure of association, either relative risk or population attributable fraction, when calculating power and sample size.

Mortality associated with diastolic hypertension and isolated systolic hypertension among men screened for the Multiple Risk Factor Intervention Trial

The large cohort of white men (317,871) 35 to 57 years old at initial screening for possible enrollment into the Multiple Risk Factor Intervention Trial (MRFIT) was examined with regard to initial blood pressure levels and subsequent coronary heart disease (CHD), stroke, and all-cause mortality. The overall prevalence of isolated systolic hypertension (ISH), defined as systolic blood pressure (SBP) greater than or equal to 160 mm Hg and diastolic blood pressure (DBP) less than 90 mm Hg, was 0.67% among white men screened for MRFIT and increased with age (0.31% among 35- to 39-year-olds to 1.7% among 55- to 57-year-olds). The 6 year CHD and all-cause mortality rates in men over 50 were highest in those with ISH compared with both subjects with diastolic hypertension and those with normal pressure. The relative risk of death from stroke in those with ISH, compared with that in those with SBP less than 160 mm Hg and those with DBP less than 90 mm Hg, was 3.0 (95% confidence interval 1.3 to 6.8). In addition, at any level of DBP, the level of SBP appeared to be the major determinant of all-cause and CHD mortality. The determinants of ISH in individuals under 60 years of age as well as the possible efficacy of its treatment should be evaluated further.

An estimate of the proportion of colo-rectal and stomach cancers which might be prevented by certain changes in dietary habits

The colo-rectum and stomach are the 2 sites for which the role of diet in the etiology of cancer is most heavily implicated. Quantitative estimates as to the proportion of cancer deaths attributable to diet vary considerably. This may be partially due to the application of the concept of population attributable risk which has proved a useful yardstick in estimating the public health impact of completely removing an exposure factor under study. This simple concept may well serve in occupational epidemiology, but with more and more risk factors found on a metric scale, or at least characterized on an ordinal scale, extensions should be considered. In this report, a preventable proportion is defined as the excess risk which would be removed if the population exposure distribution were to change ina favorable direction. A simple way of modelling such changes is proposed. This new measure is illustrated with data from 6 case-control studies on colo-rectal or stomach cancer. The results indicate that the proportion of these cancers which may be prevented assuming arbitrary shifts of a single risk factor towards the next lower risk category if populations were to change their dietary habits is in the order of magnitude of 15% to 20%. This figure is much lower, but not necessarily in contradiction to the figures considered for the proportion of cancers attributable to diet.

Cardiac involvement in essential hypertension. Prevalence, pathophysiology, and prognostic implications

The understanding of the complex involvement of the heart in human hypertension has been substantially advanced by the development of methods that detect left ventricular hypertrophy. This article reviews the literature regarding cardiac involvement in systemic hypertension, blood pressure relations to cardiac hypertrophy or clinical prognosis, relation between cardiac structure and cardiac performance in hypertensive patients, and finally, neurohormonal influence on the activity of the heart.

Incidence and precursors of hypertension in young adults: the Framingham Offspring Study

The occurrence of hypertension and its precursors is examined in the Framingham Offspring Study of 2,027 men and 2,267 women ages 20-49 years followed for 8 years. The age-specific prevalence of hypertension was similar at both the first (1971-1975) and the second (1979-1983) examination for both men and women. Prevalence rates were higher among men than among women, and there was a higher rate of hypertension treatment at the second exam, particularly among women, 75% of whom reported being treated for hypertension. The incidence of hypertension in participants free from hypertension at the first examination increased threefold from the second to the fifth age decades in men and eight-fold in women. Under age 40, men were twice as likely as women to develop hypertension, but after age 40, 8-year incidence rates were similar in men (14.2%) and women (12.9%). Adiposity, relative weight, heart rate, alcohol intake, hematocrit, blood sugar, serum protein, triglyceride, and phosphorous were all related to hypertension occurrence in one or both sexes, controlling for age. In multivariate analysis, adiposity (P less than 0.01), heart rate (P less than 0.01), and triglyceride (P less than 0.05) were all significant independent predictors of hypertension in men. In women, adiposity (P less than 0.001), heart rate (P less than 0.01), hematocrit (P less than 0.05), and alcohol consumption (P less than 0.05) were independent contributors. When controlling for blood pressure measured at the first examination, the best single predictor of hypertension incidence, the multivariate assessment changed very little. Adiposity stands out as a major controllable contributor to hypertension. Changes in body fat over 8 years were related to changes in both systolic and diastolic blood pressure. Markedly obese women in their fourth decade were seven times more likely to develop hypertension than were lean women of the same age. Weight control deserves a high priority in efforts to prevent hypertension in the general population.

Left ventricular hypertrophy in hypertension. Prevalence and relationship to pathophysiologic variables

In less than a decade since development of echocardiographic measurement of left ventricular muscle mass, studies using this technique have provided considerable information about the prevalence and pathophysiology of left ventricular hypertrophy in human hypertension. Increased left ventricular mass has been found in a significant minority of patients with systemic hypertension, with the exact prevalence dependent both on how a population is selected and on the sex, race, and possibly age composition of its members. All published studies have reported that left ventricular hypertrophy is more closely related to blood pressure recorded in the patient's natural setting during normal activity or exercise-whether measured by portable recorder or home manometer-than to blood pressure measured by the physician. In addition, studies indicate that the classic hypertensive abnormalities of concentric left ventricular hypertrophy and increased peripheral resistance are interrelated, while left ventricular hypertrophy is absent in a subgroup of patients with mild essential hypertension who exhibit high cardiac output and evidence of supernormal myocardial contractility. Conversely, the left ventricular functional response to exercise is inversely related to the degree of hypertrophy. High levels of blood viscosity, which would tend to blunt the reduction in peripheral resistance expected during sleep or exercise, have also been associated with left ventricular hypertrophy in patients with essential hypertension. Echocardiographic studies have provided evidence both for and against the hypothesis that activity of the sympathetic or reninangiotensin systems plays a direct role in causing hypertensive cardiac hypertrophy. These findings demonstrate the useful role that echocardiographic assessment of left ventricular structure and function may play in hypertension research.

Need to prevent and control high-normal and high blood pressure, particularly so-called "mild" hypertension: epidemiological and clinical data

The need to prevent and control high blood pressure (HBP), including so-called "mild" hypertension [diastolic blood pressure (DBP) 90-104 mm Hg in adults age 30+] stems from the extensive data on the increased risks due to these common blood pressure (BP) levels, including risk of catastrophic cardiovascular events (coronary, cerebrovascular, etc.), both nonfatal and fatal. Prospective population data from the national cooperative Pooling Project and the Chicago Heart Association Detection Project in Industry illustrate the extensively documented facts. They also show that only a small minority of middle-aged and older Americans have optimal low-normal BP levels, i.e., DBP less than 80 mm Hg (SBP less than 120). Thus, the problem of BP above optimal level for health over a long life span is a population-wide problem. The data also show that the great majority of excess catastrophic events attributable to elevated BP occur among people with DBP 90-104 and 80-89 mm Hg, levels very common in the population. Most people with such BP levels also have one or more other major risk factors (e.g., hypercholesterolemia, cigarette use, ECG abnormalities) and thus are at markedly increased risk, both relative and absolute. In addition to these excess risks for major illness, disability, and death, people with BP above optimal levels are more highly prone to other events, clinical and subclinical, that have adverse effects on long-term prognosis, including development of target organ damage and severe hypertension. These data lead to the following inferences about medical care and public health strategy: (a) A key task is, by safe nutritional-hygienic means, to shift the entire population distribution of BP downward, for both primary and secondary prevention of HBP. Such means include prevention and control of obesity, high sodium and alcohol intake, and sedentary habit, from early childhood on. (b) People with DBP 80-89 mm Hg need to be identified promptly, with institution of nutritional-hygienic measures to prevent development of frank hypertension and to correct other risk factors. (c) People with DBP 90-104 and higher need to be identified promptly, with institution of measures to normalize BP and control other major risk factors, by nutritional-hygienic means alone whenever possible or in combination with drug treatment for HBP when necessary to prevent organ system damage, serious illness, disability, and premature death.(ABSTRACT TRUNCATED AT 400 WORDS)

Coronary artery disease in blacks: risk factors

The current literature indicates that of the major risk factors for coronary artery disease (CAD), United States blacks and whites have similar rates for cigarette smoking and cholesterol levels. The prevalence of diabetes mellitus is higher in black females than white females. Both black males and females have higher prevalence rates for hypertension. These differences in risk factors between blacks and whites in spite of similar degrees of CAD suggest that the relative importance of specific risk factors might differ between the two racial groups. Research is needed to determine if there are protective factors in blacks (e.g., high-density lipoprotein cholesterol) and/or previously unrecognized risk factors (e.g., diuretic-induced lipid abnormalities) that may be playing a major role in the epidemiology of CAD in the black population.

Serum cholesterol, other risk factors, and cardiovascular disease in a Japanese cohort

The relationship of serum cholesterol and other risk factors to cardiovascular disease was investigated in a 16-year cohort of 16,711 residents of Hiroshima and Nagasaki. Examined in detail were the relationship of serum cholesterol, and the joint relationships of serum cholesterol, systolic blood pressure, diastolic blood pressure, and other risk factors to coronary heart disease (CHD), cerebral infarction (CI), and cerebral hemorrhage (CH). Baseline and biennially collected risk factor data were analyzed. The latter type of measurement permitted separate investigation of both the short-term and long-term effects of cholesterol measurements. In both types of analyses, both serum cholesterol and blood pressure showed strong associations with CHD incidence. In particular, there were strong associations with short-term and delayed CHD incidence. Furthermore, the association of cholesterol with short-term CHD incidence could not be explained by its association with delayed CHD incidence, or vice versa. Multivariate analyses that also included several other risk factors (smoking habits, clinical diagnosis of diabetes mellitus, left ventricular hypertrophy or strain on electrocardiogram, relative body weight, hematocrit, and proteinuria) for which data were available showed such risk factors to be of lesser, but generally non-negligible, importance in this population. In the case of CH and CI, serum cholesterol was found to be weakly or not at all related to incidence of either disease while blood pressure remained a strong correlate. For CI some suggestion of a statistical interaction between blood pressure and serum cholesterol was found. Discussed are implications for theories of disease pathogenesis for CHD, CI and CH.