Showing papers in "American Journal of Epidemiology in 1986"

Total energy intake: implications for epidemiologic analyses

Abstract: Associations between intake of specific nutrients and disease cannot be considered primary effects of diet if they are simply the result of differences between cases and noncases in body size, physical activity, and metabolic efficiency. Epidemiologic studies of diet and disease should therefore be directed at the effect of nutrient intakes independent of total caloric intake in most instances. This is not accomplished with nutrient density measures of dietary intake but can be achieved by employing nutrient intakes adjusted for caloric intake by regression analysis. While pitfalls in the manipulation and interpretation of energy intake data in epidemiologic studies have been emphasized, these considerations also highlight the usefulness of obtaining a measurement of total caloric intake. For instance, if a questionnaire obtained information on only cholesterol intake in a study of coronary heart disease, it is possible that no association with disease would be found even if a real positive effect of a high cholesterol diet existed, since the caloric intake of cases is likely to be less than that of noncases. Such a finding could be appropriately interpreted if an estimate of total caloric intake were available. The relationships between dietary factors and disease are complex. Even with carefully collected measures of intake, consideration of the biologic implications of various analytic approaches is needed to avoid misleading conclusions.

A data-based approach to diet questionnaire design and testing

Abstract: A self-administered diet history questionnaire has been developed for epidemiologic and clinical use. Both the food list and the nutrient values to be associated with it were developed using dietary data from 11,658 adult respondents to the Second National Health and Nutrition Examination Survey (NHANES II). Food items were selected on the basis of their contribution to total population intake of energy and each of 17 nutrients in the NHANES II data, and represent over 90% of each of those nutrients. Associated nutrient composition values were determined from the NHANES II database using frequency of consumption data in that survey. Portion sizes to be associated with each food item were derived from observed portion size distributions in NHANES II, based on three-dimensional models. The resulting food list and its corresponding brief data base, when used to calculate nutrients from a diet record, yielded correlations of r greater than 0.70 with the more detailed method. Field administration produced mean values comparable to national data.

The diet and 15-year death rate in the seven countries study.

Abstract: In 15 cohorts of the Seven Countries Study comprising 11579 men aged 40-59 years and healthy at entry 2288 died in 15 years. Death rates differed among cohorts. Differences in mean age blood pressure serum cholesterol and smoking habits explained 46% of variance in death rate from all causes 80% from coronary heart disease 35% from cancer and 45% from stroke....Death rates were related positively to average percentage of dietary energy from saturated fatty acids negatively to dietary energy percentage from monounsaturated fatty acids and were unrelated to dietary energy percentage from polyunsaturated fatty acids proteins carbohydrates and alcohol. Inclusion of the ratio of monounsaturated to saturated fatty acids "with age blood pressure serum cholesterol and smoking habits as independent variables accounted for 85% of variance in rates of deaths from all causes 96% coronary heart disease 55% cancer and 66% stroke....All-cause and coronary heart disease death rates were low in cohorts with olive oil as the main fat. Causal relationships are not claimed but consideration of characteristics of populations as well as of individuals within populations is urged in evaluating risks." The study covered subpopulations in the United States Finland the Netherlands Italy Yugoslavia Greece and Japan. (EXCERPT)

Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women

Abstract: To assess the validity of self-reported illnesses, medical records were reviewed for participants reporting major illnesses on the biennial follow-up questionnaires used in a prospective cohort study which began in 1976. In over 90% of cases of cancer of the breast, skin, large bowel, and thyroid, histopathology reports confirmed the subjects' self-report. Lower levels of confirmation were obtained for cancers of the lung, ovary, and uterus. Application of strict diagnostic criteria also gave lower levels of confirmation for myocardial infarction (68%) and stroke (66%). Among random samples of women reporting fractures and hypertension all records obtained confirmed self-reports. For self-reported elevated cholesterol levels 85.7% of self-reports were confirmed. Self-report is a valuable epidemiologic tool but may require additional documentation when the disease is diagnostically complex.

Depressive symptoms in relation to physical health and functioning in the elderly

Abstract: The associations between depressive symptoms and functional disability and chronic conditions are examined in an elderly cohort of 2,806 noninstitutionalized men and women living in New Haven, Connecticut who were interviewed in 1982 as a part of the Yale Health and Aging Project. The aim is to explore several potential sources of invalidity in using the Center for Epidemiologic Studies-Depression scale (CES-D) to measure depressive symptoms in elderly populations. In particular, the authors are concerned with the possibility that prevalent physical illnesses and disabilities may cause the older person to report many somatic complaints, a major component of most measures of depressive symptomatology, and thereby inflate his or her CES-D score. Mean CES-D scores are 4.86 for those without any disabilities and range to 13.51 for those with major functional disabilities. However, physical disability is significantly associated with virtually every item on the CES-D scale not just those somatically-oriented items. The addition of functional disability to a multivariate model including age subfactor analysis of responses from this elderly sample produces results almost identical to those reported by earlier investigators who studied younger and middle-aged adults. The authors conclude that physical disabilities among the elderly do not appear to be a major threat to the validity of the CES-D scale and that the strong associations between physical and mental health should be rigorously investigated.

Binomial regression in glim: estimating risk ratios and risk differences

Abstract: Although an estimate of the odds ratio adjusted for other covariates can be obtained by logistic regression, until now there has been no simple way to estimate other interesting parameters such as the risk ratio and risk difference multivariately for prospective binomial data. These parameters can be estimated in the generalized linear model framework by choosing different link functions or transformations of binomial or binary data. Macros for use with the program GLIM provide a simple method to compute parameters other than the odds ratio while adjusting for confounding factors. A data set presented previously is used as an example.

Social ties and mortality in evans county, georgia

Abstract: In an attempt to replicate Berkman and Syme's study of social networks and mortality in Alameda County, California, the authors investigated the relationship between a social network index and survivorship from 1967 to 1980 in the Evans County, Georgia, cohort. They constructed an index modeled after the Berkman Social Network Index and tested it in race- and sex-specific proportional hazards models for 2,059 subjects who were examined in 1967-1969 during the Evans County Cardiovascular Epidemiologic Study. The present study emphasized a priori specification of the social network index and statistical hypothesis test. Descriptive analyses were consistent with a modest social networks effect (e.g., hazard ratio (95 per cent confidence interval) of 1.6 (1.2-2.2) ). Among white males, the age-adjusted hazard ratio comparing the lowest to the highest value of our six-level index was 2.0 (1.2-3.4), but control for potential confounders (principally cardiovascular disease risk factors) reduced this value to 1.5 (0.8-2.6). The social networks effect among white females, black males, and black females was weaker and clearly nonsignificant. Exploratory analyses suggested that marital status, church activities, and an alternate social network index predicted survivorship, but not in a dose-response fashion. Reduced survivorship among older subjects with few social ties was the most important feature of the data.

Coronary heart disease mortality and alcohol consumption in framingham

Abstract: The relationship between ethanol consumption and coronary heart disease was examined in the original Framingham Heart Study cohort (1948) with a 24-year follow-up from exam 2 (2,106 males and 2,639 females). Ethanol consumption shows a strong U-shaped relationship with coronary heart disease mortality for male nonsmokers and heavy smokers both in the raw age-adjusted data and in the Cox regression analyses, where ethanol consumption is modeled quadratically. No ethanol effects were found for female nonsmokers. The age-adjusted data suggest a U-shape curve for female smokers, although this was not confirmed by the Cox analysis. Separate analyses relating alcohol consumption to mortality from all causes showed similar effects except that the reduction in mortality for males was much less. For male coronary heart disease mortality, ethanol consumption was subdivided into beer, wine, and spirits consumption. These beverages were also modeled quadratically in the Cox analyses, and all showed strong U-shaped curves for both nonsmokers and heavy smokers. In nonsmokers, beer and wine show greater reductions in coronary heart disease mortality than spirits.

Use of time to pregnancy to study environmental exposures

Abstract: There is need in reproductive epidemiology for sensitive and convenient screening tools that can be used to study environmental and occupational exposures. The measurement of fecundability (the probability of pregnancy in each cycle) by ascertaining how long it takes couples to conceive, may be useful for this purpose. Theoretically, exposures that interfere with any of the biologic processes involved in achieving pregnancy could lower fecundability among exposed men or women. To evaluate problems with collecting data on time to pregnancy, telephone interviews were conducted with nearly 700 pregnant women who reported having planned their pregnancies. Power curves were developed based on the distribution of time to pregnancy in the interviewed population. These curves indicate that relatively small sample sizes are sufficient for investigating an exposure. For example, the authors estimate that to detect a given 50% drop in mean fecundability with 80% power would require data from 55 exposed and 55 unexposed women who are pregnant. Disadvantages of using time to pregnancy as a reproductive endpoint include susceptibility to selection bias and need for data on several potential confounding variables. The next step in evaluating time to pregnancy as a reproductive endpoint is to apply it in studies of environmental or occupational exposures.

Relationship of clinical diabetes and asymptomatic hyperglycemia to risk of coronary heart disease mortality in men and women.

Abstract: This epidemiologic study explored the sex differential in risk of death from coronary heart disease in persons with or without clinically diagnosed diabetes or asymptomatic hyperglycemia. Use was made of 9-year follow-up data from the Chicago Heart Association Detection Project in Industry for 11,220 white men and 8,030 white women aged 35-64 years at entry to the Project (November 1967-January 1973). Both clinically diagnosed diabetes and asymptomatic hyperglycemia were associated with an increased risk of death from coronary heart disease. The extent of this association was greater in women than in men in regard to relative risk. However, absolute excess risk for both diabetics and those with asymptomatic hyperglycemia was larger for men than for women. Clinical diabetes appeared to be an independent risk factor for coronary heart disease in both men and women based on multivariate Cox regression analyses. On the other hand, for men, no significant independent effect of asymptomatic hyperglycemia was apparent. Women with asymptomatic hyperglycemia had significantly higher coronary heart disease death rates than normoglycemic women, with adjustment for major coronary heart disease risk factors; in multivariate analyses, the relationship of asymptomatic hyperglycemia to risk of coronary heart disease was of borderline significance (p = 0.054). This study indicates the independent associations of diabetes and possibly asymptomatic hyperglycemia with coronary heart disease mortality, with greater relative significance in women than in men.

Osteoporosis and the risk of hip fracture

Abstract: The incidence of cervical and intertrochanteric proximal femur fractures at various levels of cervical and intertrochanteric bone mineral density, respectively, was estimated by using population-based data from ongoing studies of osteoporosis and fractures among women residing in Rochester, Minnesota. Hip fractures were uncommon among women with femoral bone density greater than or equal to 1.0 g/cm2, but their frequency increased as bone density declined below that point at both femoral sites. The incidence of cervical femur fractures was estimated at 8.3 per 1,000 person-years among women with cervical bone density less than 0.6 g/cm2, while the estimated incidence of intertrochanteric femur fractures reached 16.6 per 1,000 person-years among those with intertrochanteric bone density less than 0.6 g/cm2. This new approach to the assessment of fracture risk from bone mineral density measurements indicates that osteoporosis is an important underlying cause of hip fractures.

Analytical potential for multiple cause-of-death data

Abstract: This paper describes multiple cause-of-death data as produced by the [U.S.] National Center for Health Statistics....As background for understanding the nature of multiple cause-of-death data a discussion of the medical certification on the death certificate is presented. The epidemiologic potential of multiple cause data is then explored through examples which use the newly available data and through a review of past and current applications of multiple cause data. Some further comments on problems in death certification by George W. Comstock and Robert E. Markush (pp. 180-1) are also included. (EXCERPT)

A general estimator for the variance of the mantel haenszel odds ratio

Abstract: This commentary presents a new estiof matching variables, unmatched varimator of the variance of the natural log of ables, or both. This notation encompasses the Mantel-Haenszel odds ratio and illusthe full spectrum of matched and untrates its application in typical case-control matched designs. For example, in the pairdesigns. In contrast with any single previmatched study, MI and Mo are always one; ously proposed estimator (1-7), this estiin a design with J controls uniquely mator is easily computed and can be used matched to each case (1: J matching), MI in the analysis of data from any sampling is always one and Mo is always J; in an design (e.g., individually matched, category unmatched design, MI and Mo will usually matched, or unmatched). Furthermore, in vary in an arbitrary fashion across the a series of Monte Carlo simulations, this strata. estimator yielded confidence limits as good Suppose now that the parameter of inas or better than those obtained using preterest is the odds ratio and the odds ratio vious estimators (8). has a common (constant) value across strata. If R = AD/T and S = BC/T, the BACKGROUND Mantel-Haenszel estirnator2f the common We will use the following notation to odds ratio is defined as ORMH = R+/S+, where R+ = C, R, S+ = C, S, and the sumrepresent data values in a basic stratum mations are acrzss all strata. from an epidemiologic study: Valid use of O R M ~ requires that the study ' Exposed Unexposed Totals sample be \"large,\" but there are at least two

Risk factors for nosocomial urinary tract infection

Abstract: In a prospective study carried out in June 1979-April 1981, 134 of 1,458 adult inpatients at New England Deaconess Hospital, Boston, Massachusetts, acquired 136 urinary tract infections during 1,474 indwelling bladder catheterizations. Multiple logistic regression analysis identified nine factors that were significantly associated with acquisition of infection: duration of catheterization, lack of systemic antibiotic during short catheter courses, lack of urinemeter drainage, female sex, diabetes mellitus, microbial colonization of the drainage bag, serum creatinine greater than 2 mg/dl at the time of catheterization, the reason for catheterization, and the use of catheters with sealed collection junctions when no antibiotic was administered. When potential risk factors were considered individually, without adjusting for the influence of other factors, infection was also significantly associated with 11 other factors, including several that have been previously reported as risk factors for catheter-associated urinary tract infections; however, these associations were no longer significant after adjustment for one or more of the nine factors noted above. These findings suggest approaches to surveillance, prevention, and research activities. They also strongly support the widely-held but poorly documented belief that persons with diabetes are more susceptible to urinary tract infection than are persons without diabetes.

Existence of consistent hypo- and hyperresponders to dietary cholesterol in man

Abstract: Hyper- and hyporesponsiveness of serum cholesterol to dietary cholesterol is an established concept in animals but not in man. The authors studied the stability of the individual response of serum cholesterol to dietary cholesterol in three controlled experiments in 1982. The subjects were volunteers from the general population living in or near Wageningen, the Netherlands. Each experiment had a low-cholesterol baseline period (121, 106, and 129 mg/day in experiments 1, 2, and 3, respectively) and a high-cholesterol test period (625, 673, and 989 mg/day). Duplicate portion analysis showed that dietary cholesterol was the only variable. The 94 healthy men and women who completed experiment 1 showed an increase (mean +/- standard deviation (SD] in serum cholesterol of 0.50 +/- 0.39 mmol/liter (19 +/- 15 mg/dl). Seventeen putative hyperresponders, defined by their response in experiment 1, were retested in experiments 2 and 3; they showed responses of 0.28 +/- 0.38 mmol/liter (11 +/- 15 mg/dl) and 0.82 +/- 0.35 mmol/liter (32 +/- 14 mg/dl), respectively. Fifteen hyporesponders, selected in experiment 1, showed responses in experiments 2 and 3 of 0.06 +/- 0.35 mmol/liter (2 +/- 14 mg/dl) and 0.47 +/- 0.26 mmol/liter (18 +/- 10 mg/dl), significantly lower than the corresponding values for hyperresponders. The standardized regression coefficient for individual responses in experiment 2 on those in experiment 1 was beta = 0.34 (p = 0.03, n = 32); the corresponding regression coefficient for experiment 3 and experiment 1 was 0.53 (p less than 0.01). After correction for intraindividual fluctuations the true responsiveness distribution was found to have a between-subject standard deviation of about 0.29 mmol/liter (11 mg/dl). This implies that if the mean response to a certain dietary cholesterol load amounts to e.g., 0.58 mmol/liter (22 mg/dl), then the 16% of subjects least susceptible to diet will experience a rise of only 0.29 mmol/liter (11 mg/dl) or less, while in the 16% of subjects most susceptible to diet, serum cholesterol will rise by 0.87 mmol/liter (34 mg/dl) or more. The authors conclude that modest differences in responsiveness of serum cholesterol to dietary cholesterol do exist in man, and that the wide scatter of responses observed in single experiments is largely due to chance fluctuations.

Epidemiologic associations with nuclear, cortical, and posterior subcapsular cataracts

Abstract: Data from the 1971-1972 National Health and Nutrition Examination Survey were used to examine the associations between specific cataract types and age, race, sex, number of years spent in school, diabetes, systolic blood pressure, urban versus rural residence, and average daily ultraviolet spectrum B (UV-B) radiation counts at the site of examination. Separate comparisons were made of persons with no lens changes (n = 1,299) and persons with cortical (n = 55), nuclear (n = 104), and posterior subcapsular cataracts (n = 18). Persons with more than one type of cataract were excluded from the analysis. Cortical cataracts were associated with age (relative risk (RR) = 11.4 for age 70 years vs. age 50 years), race (RR = 3.5 for blacks vs. whites), sex (RR = 3.0 for women vs. men), educational achievement (RR = 1.8 for less than nine years of schooling vs. college), and UV-B count (RR = 3.6 for 6.0 X 10(3) counts vs. (2.6 X 10(3) counts). Nuclear cataracts were associated with age (RR = 38.6 for age 70 years vs. age 50 years), race (RR = 1.8 for black vs. white), and residence (RR = 1.6 for rural vs. urban). Posterior subcapsular cataracts were associated with diabetes (RR = 6.6 for diabetes present vs. diabetes absent) and systolic blood pressure (RR = 2.2 for 160 mmHg vs. 120 mmHg). Cortical cataracts were more common in women and more often found in locations with increased UV-B radiation counts than either nuclear or posterior subcapsular cataracts. In diabetics, the risk of posterior subcapsular cataracts was greater than the risk of nuclear cataracts (p less than 0.05) and also appeared to be greater than the risk of cortical cataracts (p = 0.06).

Diet and pancreatic cancer: a case-control study

Abstract: In a population-based case-control study carried out in Sweden in 1982-1984, the authors examined the association of pancreatic cancer with several dietary factors, coffee, alcohol, and tobacco. Analyses were based on 99 cases, 138 population controls, and 163 hospital controls. The cases were persons aged 40-79 years diagnosed with cancer of the exocrine pancreas at three surgical departments in Stockholm and Uppsala. The risk increased with higher consumption frequency of fried and grilled meat in the comparison with each series of controls (e.g., relative risk (RR) = 1.7 (90% confidence interval (CI) = 1.1-2.7) for weekly intake and RR = 13.4 (90% CI = 2.4-74.7) for almost daily intake, in the comparison with population controls). Furthermore, associations were found with other fried or grilled foods, but not with meat other than fried or grilled. The risk also increased with the intake of margarine (e.g., RR = 9.7 (90% CI = 3.1-30.2) for 15+ g of margarine on a slice of bread, in the comparison with population controls). In contrast, no excess risk was associated with high intake of butter. A low risk was associated with frequent consumption of fruits and vegetables, particularly carrots (RR = 0.3 (90% CI = 0.2-0.7)) and citrus fruits (RR = 0.5 (90% CI = 0.3-0.9)) for almost daily intake. No consistent associations were found with coffee, artificial sweeteners or alcohol consumption, but a threefold increase in risk was associated with smoking at least one pack of cigarettes per day.

The effects of exposure misclassification on estimates of relative risk.

Abstract: In epidemiologic studies, individuals may be misclassified with respect to exposure to a risk factor for disease. Such misclassification causes the relative risk of disease associated with the exposure in the population to be biased toward the null value. Here, a formula is derived for the apparent relative risk under misclassification (R) as a function of the sensitivity (U) and specificity (V) of the test for exposure and of the true relative risk (R) and true prevalence of exposure (P(E] in the population. The relative influence of U and V on the bias in R depends both on R and on P(E), with U tending to be more influential at higher values of P(E). When there is misclassification of exposure, variation in P(E) may bias comparisons of relative risk between groups or exposures, either by producing spurious differences or by masking true differences, and may generate spurious trends associated with a third variable such as age. Because the possible effects of misclassification of exposure on relative risk are complex and not easily generalized, the potential degree of bias should be evaluated individually in each situation.

A prospective study of parental history of myocardial infarction and coronary heart disease in women

Abstract: Among 121,964 women aged 30-55 years in 1976, 117,156 who were initially free from coronary heart disease provided information on a number of coronary risk factors including parental history of myocardial infarction and were followed prospectively. In 1976, 31,101 (26.5%) reported that at least one parent had suffered a myocardial infarction. Questionnaires in 1978 and 1980 identified women who had developed nonfatal myocardial infarction (n = 132) and angina pectoris (n = 101). Fatal coronary heart disease cases (n = 42) were ascertained by searches of state vital records. The age-adjusted relative risk of nonfatal myocardial infarction for women with a parental history of myocardial infarction less than or equal to 60 years of age compared with women with no family history was 2.8 (95% confidence limits (CL) 1.8, 4.3). For those with a parental history of myocardial infarction greater than 60 years of age, the age-adjusted relative risk of nonfatal myocardial infarction was 1.0 (CL 0.5, 1.8). The age-adjusted relative risks of fatal coronary heart disease were 5.0 (CL 2.7, 9.2) for parental history before age 61 and 2.6 (CL 1.1, 5.8) for parental history after age 60. The corresponding relative risks of angina pectoris were 3.4 (CL 2.2, 5.2) and 1.9 (CL 1.2, 3.2), respectively. These associations were only slightly altered by adjustment for history of hypertension, diabetes, high cholesterol, use of oral contraceptives, menopause, postmenopausal hormone use, obesity, or smoking, in individual stratified analysis or in multivariate analyses. These data support the hypothesis that parental history of myocardial infarction has an independent effect on risk that is not explained solely by individual risk factors.

Does breastfeeding really save lives, or are apparent benefits due to biases?

Abstract: Analysis of mothers recall data collected in 1976-77 by a probability survey in Peninsular Malaysia shows an association between breastfeeding up to 6 months of age and improved survival of infants throughout the 1st year of life. Inappropriate sample selection and inadequate control of confounding can introduce large biases in these analyses. The magnitude and direction of these biases are presented. Even when these biases are dealt with unsupplemented breastfeeding appears more beneficial than supplemented breastfeeding. The younger the infant and the longer the breastfeeding the greater the estimated benefits in terms of deaths averted. The use of powdered infant formula did not appear to offset the detrimental effects of early weaning and supplementation. The positive relationships found in these analyses between breastfeeding and survival are not due to death precluding or terminating breastfeeding. Nor are they likely to be due to a shift away from breastfeeding because of recent illness which was also controlled in the analyses. Nor are they likely to be due to other factors that both increase mortality risk and shorten breastfeeding; when such factors are taken into account the beneficial effects of breastfeeding become stronger and imply that if there had been no breastfeeding in this sample twice as many babies would have died after the 1st week of life. (authors)

Association of low birth weight with passive smoke exposure in pregnancy

Abstract: In a prospective study of 3,891 antenatal patients at Yale-New Haven Hospital between 1980 and 1982, one fourth (23.6%) had not smoked cigarettes during pregnancy but had been exposed to sidestream smoke for at least two hours per day. Among the nonsmokers, passive smoke exposure was significantly related to delivering a low birth weight (less than 2,500 g) newborn. This relation only occurred in term (greater than or equal to 37 weeks) deliveries. Compared with unexposed women, the relative risk of low birth weight after adjustment for confounding factors was 2.17 (95% confidence interval (CI) = 1.05-4.50). Those exposed to passive smoke delivered infants 24 g lighter on average. There was no additive effect of passive smoking on smokers themselves. Repeating the analysis on all women with term deliveries, therefore, resulted in a slightly diminished risk of low birth weight due to passive smoking of 1.52 (95% CI = 0.90-2.56). The risk of low birth weight at term due to direct cigarette smoking was 3.54 (95% CI = 1.62-7.71). Gestational age was unrelated to passive smoking, which appears to exert its effect primarily through growth retardation in term newborns.

Sedentary jobs and colon cancer

Abstract: The association between physical job activity and colon cancer was examined in a 19-year follow-up study of 1.1 million Swedish men. The relative risk (RR) of colon cancer in men employed in sedentary occupations was estimated at 1.3 (1.2-1.5, 90% confidence interval), with the highest risk for the transverse colon including flexures (RR = 1.6) and the lowest risk for the sigmoid (RR = 1.2). The relative risk for rectal cancer was not elevated. Age, population density, and social class did not confound the relative risk and were controlled for, whereas marital status and geographic region did not. Furthermore, food habits and physical activity during leisure time were examined in the same occupations in another sample and it was judged unlikely that confounding from these factors could explain the association.

A comparison of three indicators for identifying Mexican Americans in epidemiologic research. Methodological findings from the San Antonio Heart Study.

Abstract: Because the issue of how to empirically identify Mexican Americans in health-related research is still unresolved, the authors compared the performance of three indicators for identifying Mexican Americans across five distinct population subgroups: men and women in two age strata, and residents in low, middle, and high socioeconomic neighborhoods. Individual surname had the lowest sensitivity, specificity, and predictive values in the pooled population sample and varied the most widely on these parameters across population subgroups. Parental surnames, which are available on vital statistics and could easily be added to other health records used in secondary analyses, offered a significant improvement over individual surname in classifying persons as Mexican American. The San Antonio Heart Study (SAHS) algorithm, a nine-item indicator which uses parental surnames, birthplace of both parents, self-declared ethnic identity, and ethnic background of grandparents, had the highest sensitivity, specificity, and predictive values and varied the least on these parameters across different sex, age, and socioeconomic status population subgroups. The performance of all indicators was lower at the higher socioeconomic status levels. The findings suggest that it may be useful to use parental surnames as an indicator for Mexican-American ethnicity in research involving vital statistics and to add parental surnames to other health records frequently used in secondary analyses. Since the SAHS algorithm can be adapted for use with non-Mexican origin Hispanic subgroups, it may be a useful indicator for Mexican-American (or other Hispanic) ethnicity in survey research.

Validity of the fatty acid composition of subcutaneous fat tissue microbiopsies as an estimate of the long-term average fatty acid composition of the diet of separate individuals

Abstract: The relationship between the fatty acid composition of subcutaneous adipose tissue and diet was estimated in 59 Dutch women aged 32-35 years. Food consumption was estimated by taking the means of nineteen 24-hour recalls administered over a period of two and a half years, August 1981-December 1983. Highly significant correlations were found between linoleic acid content of fat tissue and diet (r = 0.70) and also between the linoleic acid-to-saturated fatty acid (linoleic/S) ratio of fat tissue and diet (r = 0.62). This confirms the hypothesis that on an individual level the fatty acid composition of the adipose tissue is a valid index for the habitual dietary fatty acid composition of free-living adults. When using one 24-hour recall instead of the average of 19 recalls, the correlation coefficient between the linoleic/S ratio of the diet and that of the adipose tissue was substantially decreased. This demonstrates the weakening effect of the large day-to-day variation in within-person intake on the correlation between a short-term assessment of the nutrient intake of an individual and a biochemical indicator of long-term nutritional status.

A case-control study of amyotrophic lateral sclerosis.

Abstract: The authors conducted a study of 518 amyotrophic lateral sclerosis patients identified between 1977 and 1979 and 518 controls to investigate putative risk factors for this disease. Occupations at risk of electrical exposure were reported more often by patients (odds ratio (OR) = 3.8, 95% confidence interval (CI) = 1.4-13.0) as were electrical shocks producing unconsciousness (OR = 2.8, 95% CI = 1.0-9.9). Although an overall excess of physical trauma associated with unconsciousness was observed in the amyotrophic lateral sclerosis patients (OR = 1.6, 95% CI = 1.0-2.4), the effect was inversely associated with duration of the unconscious episodes, suggesting an effect of recall bias. Only slight differences were found for surgical traumata to the nervous system. Parkinsonism was reported more often among first degree relatives of cases (OR = 2.7, 95% CI = 1.1-7.6). The frequencies of prior poliomyelitis or other central nervous system diseases were similar for patients and controls. Occupational exposure to selected toxic substances was similar for patients and controls except for the manufacture of plastics (OR = 3.7, 95% CI = 1.0-20.5), although few details of these exposures were provided. No differences in occupations with exposure to animal skins or hides were observed.

Effect of questionnaire design on recall of drug exposure in pregnancy

Abstract: Case-control studies of antenatal drug exposure and birth defects often rely on maternal recall of drug use in pregnancy. Differential recall among mothers of cases and controls can lead to information bias ("maternal recall bias"), and the opportunity for such bias increases as ascertainment of drug exposure diminishes. The effect of questionnaire design on the ascertainment of drug use in pregnancy was examined in two studies. In a pilot interview study of 532 women in obstetric/gynecologic practices, information on drug use in the past year was obtained by means of three questions asked in sequence: The first question was open-ended, the second asked about drug use for selected indications, and the third asked about use of specifically named drugs. Among obstetric patients who reported use of any of five drugs, less than 50% did so in response to the open-ended question, and approximately 20-40% reported use only when the specific drug name was asked. In a case-control Birth Defects Study of 5,435 mothers of malformed children, information on drug use in pregnancy was obtained by asking questions in sequence about indications and specifically named drugs. Among the women who reported use of any of 11 drugs, 6-40% did so only when asked about the specific drug by name. These findings suggest that completeness of ascertainment of antenatal drug exposure varies according to how the mother is questioned and is directly related to the specificity of the questions asked.