Measurement of von Willebrand factor activity: relative effects of ABO blood type and race.

Tests based on three different principles are reported to measure the activity of von Willebrand factor (VWF): ristocetin cofactor (VWF:RCo), collagen binding (VWF:CB), and the so-called "activity ELISA" (VWF:MoAb). We measured these and other diagnostic parameters in a population of 123 randomly selected female study controls, age 18-45 years. Type O subjects had significantly lower levels than non-O subjects in each test. Race differences were seen in all tests except VWF:RCo, with Caucasians having significantly lower levels than African-Americans. ABO differences accounted for 19% of the total variance in VWF:Ag (P < 0.0001) and race for 7% (P < 0.0001), for a total of 26%. Both effects were mediated through VWF:Ag and were independent. VWF:Ag level was the primary determinant of VWF function, accounting for approximately 60% of the variance in VWF:RCo and VWF:CB and 54% of the variance in factor VIII. The ratio VWF:RCo/VWF:Ag differed significantly by race within blood group. The median ratios were 0.97 for type O Caucasians vs. 0.79 for type O African-Americans and 0.94 for non-O Caucasians vs. 0.76 for non-O African-Americans. The ratio VWF:CB/VWF:Ag did not vary. This suggests racial differences in the interaction of VWF with GP1b but not with subendothelium. Alternatively, VWF:RCo may be regulated to maintain a relatively constant plasma level in the presence of excessive VWF:Ag. This heterogeneity within the normal population is partially responsible for the difficulty in defining diagnostic limits for von Willebrand disease.

Absence of profound hyperinsulinism in polycystic ovary syndrome is associated with subtle elevations in the plasminogen activator inhibitor system.

In order to describe potential hypofibrinolytic tendencies in young (< 35 years) polycystic ovary syndrome (PCOS) patients, we studied plasminogen activator inhibitor (PAI-1) system components in women without laboratory evidence of hyperinsulinism or hyperandrogenism. The study was a prospective, observational comparison and took place in a major urban infertility referral center. Age, body mass index, ovulatory status, selected androgen levels, fasting insulin and plasma lipids were measured in subjects with PCOS (n = 39) and normal control subjects (n = 20). Women with PCOS had higher mean serum total testosterone and androstenedione levels compared with controls (56.4 versus 40.3 ng/dl, p = 0.03, and 179 versus 133 microg/ml, p = 0.03, respectively). Mean fasting insulin levels were higher among PCOS women (p < 0.01) and were strongly correlated with PAI-1 antigen (Ag) (r = 0.46), PAI-1 activity (r = 0.43), and tissue plasminogen activator (t-PA) (r = 0.5). Correlations were evident in both PCOS and control subjects. Mean PAI-1 Ag, PAI-1 activity, and t-PA levels were significantly elevated (p = 0.003, 0.001, and 0.001, respectively) in PCOS. ANOVA was performed to control for insulin effect; a trend toward elevated PAI-1 in PCOS persisted but was no longer statistically significant (p = 0.24). PAI-1 activity elevation remained in PCOS women with mean fasting insulin levels < 10 mIU/ml (p = 0.02), yet the difference became less significant when insulin was controlled (p = 0.38). Although these data confirm known associations between insulin and PAI-1 derangements, this is the first study to quantify discrete PAI-1 elevations that persist in the setting of PCOS even with normal or low ambient insulin levels. Additional prospective studies are needed to determine whether this altered PAI-1 state is associated with a clinically important hypofibrinolytic condition and subsequent poor reproductive outcome.

Prevalence of prenatal drinking assessed at an urban public hospital and a suburban private hospital.

BACKGROUND: Although the prevalence of fetal alcohol syndrome (FAS) varies within the population, few data are available concerning variation in the prevalence of prenatal drinking. METHODS: Postpartum women delivering singleton infants at two Atlanta hospitals in 1993 or 1994 were interviewed. Those delivering infants who were small for gestational age (SGA) (n = 638) were over-sampled relative to those delivering infants with birth weights that were appropriate for gestational age (AGA) (n = 247). The prevalence of prenatal drinking was estimated as a weighted average of reports from mothers of SGA and AGA infants. Estimates of the prevalence of FAS come from the Metropolitan Atlanta Congenital Defects Program (MACDP) of the Centers for Disease Control and Prevention. RESULTS: The prevalence of first-trimester drinking was half that reported for the three previous months (private hospital: 72% vs. 35%; public hospital: 52% vs. 28%). Most women (85%) reported abstaining throughout the second trimester. Fewer than 10% of women delivering at the public hospital (7.5%), but one-quarter of those delivering at the private hospital, reported third-trimester drinking. Binge, moderate and heavy drinking in pregnancy were more common among women delivering at the public hospital. Eight infants born at the public hospital during this period, but none of those born at the private hospital, were identified as possibly having FAS; four of the eight were identified as probably having FAS. CONCLUSIONS: These results have implications for health education programs. For example, obstetricians in private practice may wish to reaffirm their advice to abstain from drinking in the third trimester. They also suggest that prenatal abstinence programs be targeted at populations identified as most likely to engage in risky drinking.

The Amblyopia Treatment Index.

OBJECTIVE: To develop a questionnaire to assess the acceptability of amblyopia treatment and its effect on the child and family. METHODS: A 20-item parental survey was developed and pilot tested on 64 subjects, aged 3 to 6 years, participating in the Amblyopia Treatment Study, a randomized trial comparing patching and atropine as treatments for moderate amblyopia. The survey was administered after 4 weeks of treatment. A descriptive item analysis and an internal consistency reliability analysis were performed. RESULTS: Nineteen of the 20 items demonstrated adequate variability as evidenced by the frequency distributions for item responses. Only 4 (<1%) of 1280 possible item responses were missing, one each by 4 different respondents. Factor analysis identified 3 treatment-related factors--"adverse effects," "compliance," and "social stigma"--among 11 of the 20 items. The internal-consistency reliability alpha for the 5-item adverse effects subscale was 0.82, the 4-item compliance subscale alpha was 0.81, and the 2-item social stigma subscale alpha was 0.84. CONCLUSIONS: The Amblyopia Treatment Index appears to be a useful instrument for assessing the impact of amblyopia treatment in 3- to 6-year-old children.

Periovulatory serum human chorionic gonadotropin (hCG) concentrations following subcutaneous and intramuscular nonrecombinant hCG use during ovulation induction: a prospective, randomized trial.

OBJECTIVE: To describe serum levels of human chorionic gonadotropin (hCG) as a function of hCG injection method (subcutaneous vs. intramuscular) among infertile women undergoing ovulation induction. DESIGN: Prospective, randomized clinical trial. SETTING: Major urban infertility referral center. PATIENT(S): Women presenting for infertility evaluation and ovulation induction. INTERVENTION(S): Controlled ovarian hyperstimulation was followed by 5,000 IU urinary (nonrecombinant) hCG injection, given intramuscularly (i.m.) or subcutaneously (s.c.). MAIN OUTCOME MEASURE(S): Serum hCG levels measured 24 hours after administration of hCG, and patient tolerability of injected hCG. RESULT(S): There were no statistically significant differences in age or body mass index (BMI) among patients receiving hCG s.c. (n = 13) or i.m. (n = 15). Mean [IQR (25; 75)] serum hCG levels in the s.c. and i.m. groups were 171.7 [27.0; 207.0] and 142.2 [102.5; 157.5] mIU/mL, respectively. No adverse events were registered by any patient receiving hCG by either injection method. In this non-IVF population, two pregnancies were established in each subgroup (4 of 28, or approximately 14% pregnancy rate). CONCLUSION(S): The s.c. administration of 5,000 IU hCG (reconstituted in vol. = 0.5 mL) was well tolerated by all women in this study and was associated with postinjection serum hCG levels similar to those observed after administration of an equivalent i.m. hCG dose. This investigation suggests that clinical use of s.c. hCG is suitable for lean women (e.g., BMI <30) undergoing ovulation induction, but additional data are needed to study the appropriateness of s.c. hCG administration in heavier patients.

Does inadequate prenatal care contribute to growth retardation among second-born African-American babies?

The authors evaluated the relation between adequacy of prenatal care and risk of delivery of full term small-for-gestational-age (SGA) infants. Data were derived from maternally linked birth certificates for 6,325 African-American women whose first two pregnancies ended in singleton, full term live births in Georgia from 1989 through 1992. The authors used stratified analysis to assess the effect of prenatal care on the risk of having an SGA baby in the second pregnancy among women with and without an SGA baby in their first pregnancy. The group of women with a history of SGA birth may be more likely to include persons for whom SGA delivery is related to factors, such as genetics, that are not amenable to intervention by prenatal care. Inadequate prenatal care was not associated with the risk of SGA delivery among women who had previously delivered an SGA baby. In unadjusted analyses, inadequate prenatal care was associated with an increased risk of delivering a full term SGA baby in the second pregnancy among women whose first baby was not SGA (risk ratio = 1.28; 95% confidence interval: 1.05, 1.55). The association did not persist when data were adjusted for confounding variables (odds ratio = 1.11; 95% confidence interval: 0.89, 1.38). Regardless of outcome in the first pregnancy, adequate prenatal care did not reduce the risk of full term SGA birth among second pregnancies in this population.

The relationship between idiopathic mental retardation and maternal smoking during pregnancy.

OBJECTIVES: Smoking has been linked to small cognitive, achievement, and behavioral deficits but has not been associated with more severe cognitive impairments. This investigation evaluated the relationship between maternal smoking during pregnancy and idiopathic mental retardation (MR). METHODS: Data on maternal smoking during pregnancy were obtained during face-to-face interviews with the mothers of 221 children with idiopathic MR and the mothers of 400 children attending public school. All children had been born in the five-county metropolitan Atlanta area in 1975 or 1976 and were living in the area when they were 10 years of age. We used exposure odds ratios (ORs) to assess the relationship between maternal smoking and MR, controlling for sex, maternal age at delivery, race, maternal education, economic status, parity, and alcohol use. RESULTS: Maternal smoking during pregnancy was associated with slightly more than a 50% increase in the prevalence of idiopathic MR (adjusted OR, 1.6; 95% confidence interval, 1.0-2.4), and children whose mothers smoked at least one pack a day during pregnancy had more than a 75% increase in the occurrence of idiopathic MR (OR, 1.9; 95% confidence interval, 1.0-3.4). This increase was neither accounted for by other sociodemographic risk factors for MR nor explained by an increase in the prevalence of low birth weight among the children of smokers. CONCLUSIONS: Our data suggest that maternal smoking may be a preventable cause of mental retardation.

The administrative prevalence of mental retardation in 10-year-old children in metropolitan Atlanta, 1985 through 1987.

OBJECTIVES: In this study, data from the Metropolitan Atlanta Developmental Disabilities Study were used to determine the administrative prevalence (i.e., the number of children previously identified for service provision) of mental retardation among 10-year-old children during the years 1985 through 1987. METHODS: Children with mental retardation (intelligence quotient [IQ] of 70 or lower) were identified by review of records from multiple sources, with the public schools as the primary source. RESULTS: The overall administrative prevalence of mental retardation was 12.0 per 1000 children. The rate for mild mental retardation (IQ of 50 to 70) was 8.4 per 1000 and the rate for severe mental retardation (IQ lower than 50) was 3.6 per 1000. The prevalence was higher in Black children than in White children (prevalence odds ratio [POR] = 2.7) and in boys than in girls (POR = 1.4). Children with severe mental retardation had more coexisting disabilities than did children with mild mental retardation. CONCLUSIONS: The mental retardation prevalence rates reported here, especially the race-specific rates, may reflect social and demographic features unique to the metropolitan Atlanta area and therefore should be used with caution in making comparisons with other populations.

Mild mental retardation in black and white children in metropolitan Atlanta: a case-control study.

OBJECTIVES: This study assessed differences in the prevalence of mild mental retardation, defined as an intelligence quotient (IQ) from 50 to 70, between Black and White children. METHODS: A case-control study design was used. Ten-year-old children with mental retardation were identified from multiple sources. Information on race, sex, maternal age, birth order, economic status, and maternal education was abstracted from birth certificates of 330 case children and 563 control children (public school students). RESULTS: The crude Black-White odds ratio (OR) was 2.6, but it was reduced to 1.8 after the other five covariates were controlled. The disparity was largest among children whose mental retardation was first diagnosed when they were 8 to 10 years old (adjusted OR = 2.5). We found no significant difference in the occurrence of mild mental retardation between Black and White children diagnosed before the age of 6 years (adjusted OR = 1.2). Black children had a higher prevalence of mild mental retardation within all strata of the other five covariates. CONCLUSIONS: Five sociodemographic factors accounted for approximately half of the excess prevalence of mild mental retardation among Black children. Possible reasons for the residual difference are discussed.

Variation in the influence of selected sociodemographic risk factors for mental retardation.

OBJECTIVES: This study explored the utility of subdividing mental retardation into groups based on the presence of other neurological conditions. METHODS: Data were abstracted from birth certificates as part of a case-control study of mental retardation among 10-year-old children. The study sample included 458 case children and 563 control children selected from public schools. Case children were subdivided on the basis of intelligence quotient (IQ) score and the presence of other neurological conditions. RESULTS: Other neurological conditions were more common with severe mental retardation than with mild mental retardation. Regardless of IQ level or the presence of other neurological conditions, boys were more likely than girls to have mental retardation. Older mothers were more likely than younger mothers to have a child with mental retardation accompanied by another neurological condition. High birth order, Black race, and low maternal education were associated with a higher prevalence of isolated mental retardation. CONCLUSIONS: These findings suggest that sociodemographic risk factors for mental retardation vary according to the presence of other neurological conditions and that subdivisions based on medical or physical criteria may be useful in epidemiologic studies of mental retardation.

Methodology to correct for differential misclassification.

Misclassification of exposure is a serious problem in epidemiology. Methods for addressing misclassification are available, but most are based on limiting assumptions such as availability of a "gold standard" measure of true exposure, or availability of two tests of exposure whose performance is nondifferential. In this paper, we discuss a method that allows the investigator to correct for differential misclassification in case-control studies. Our method only requires two potentially imperfect tests for measuring exposure. Importantly, the sensitivity and specificity of each test when applied to cases may differ from the sensitivity and specificity when applied to controls. The approach does require two subgroups of cases, such that each test's sensitivity and specificity is the same across these subgroups and requires analogous subgroups for controls. We exemplify our approach in several ways, using hypothetical data, using data from a case-control study of birth defects and service in Vietnam, and by a small Monte Carlo study. Finally, we discuss limitations of the method.

Hearing impairment among 10-year-old children: metropolitan Atlanta, 1985 through 1987.

The prevalence of hearing impairment among 10-year-old children in metropolitan Atlanta between 1985 and 1987 was evaluated. Hearing-impaired children were identified by reviewing records at public schools and health and social service agencies. The prevalence was 1.1 per 1000 and was slightly higher among Blacks and boys than among Whites and girls. The most common known causes of hearing impairment were meningitis (0.3 per 1000), genetic and hereditary conditions (0.2 per 1000), and congenital rubella syndrome (0.1 per 1000). For 55% of the children, the etiology of the hearing loss could not be determined. Most (74%) of the children were diagnosed after the age of 2, suggesting that methods of early identification need to be improved.

Prevalence of cerebral palsy among ten-year-old children in metropolitan Atlanta, 1985 through 1987.

The Metropolitan Atlanta Developmental Disabilities Study was a population-based study (1985 through 1987) to determine the prevalence of five developmental disabilities among 10-year-old children. The disabilities included cerebral palsy, mental retardation, visual impairment, hearing impairment, and epilepsy. The prevalence of cerebral palsy (CP) and a description of the children with CP are reported here. Using a record review approach, we identified 204 10-year-old children with CP (resulting in a prevalence of 2.3 per 1000). The rate of CP was significantly higher among boys (prevalence odds ratio = 1.5; 95% confidence interval = 1.1, 2.0), and the rate was also higher among black children than white children (prevalence odds ratio = 1.3; 95% confidence interval = 1.0, 1.7). Thirty-three of the children (16%) acquired CP postnatally; these children were more likely to be black or male. The gender and racial differences found for acquired CP were greater than those for congenital CP. Approximately 75% of the children had one of the other four disabilities studied; 65% of the children were mentally retarded, 46% had epilepsy, and 15% had a sensory impairment. Our multiple-source method of identifying children with CP gave us a population-based sample from which to determine the prevalence of the condition and to study factors that are associated with CP.

Mental retardation in ten-year-old children in relation to their mothers' employment during pregnancy.

We conducted a case-control study to examine relationships between potential risk factors in women's prenatal occupational histories and subsequent mental retardation in their 10-year-old children. Children with mental retardation (intelligence quotient less than 71) were identified from special education records maintained by the public school systems in the metropolitan Atlanta area and from records of various medical and social service agencies serving children with special needs. Control children were chosen from the rosters of 10-year-olds who were enrolled in regular education classes in the local public school systems. To obtain occupational histories, sociodemographic data, and other information, we interviewed 352 natural mothers (67%) of 525 case children and 408 natural mothers (64%) of 636 control children. We computed odds ratios for each of 25 selected occupation, industry, and agent categories controlling for maternal education, birth order, and race. Most comparisons yielded odds ratios that were not indicative of unusual risks, but we did find lower than expected risks among children of teachers and health-care professionals. We also found a strong, positive association between mental retardation and maternal employment in the textile and apparel industries. The findings are useful for planning the direction of future studies of childhood cognitive ability to focus on specific parental occupations or industries.

Use of two data sources to estimate odds ratios in case-control studies.

Information bias is among the most serious and common problems in epidemiology. Approaches have been developed to reduce information bias by correcting for known amounts of misclassification. Unfortunately, in most studies, the extent of exposure misclassification cannot be easily estimated. We discuss the application to case-control studies of an approach originally proposed by Hui and Walter in 1980 to estimate the sensitivity and specificity of two independent classification schemes (Hui SL, Walter SD. Biometrics 1980;36:167-171). In this paper, we propose using the EM algorithm to provide a simple numeric technique for implementing their method that seems to converge for most real-world data. Our approach allows inclusion of a measure of non-independence of the two classification schemes, and we assess the influence of non-independence on the odds ratio. Finally, we provide a simple variance estimate for the odds ratio based on the delta method and maximum likelihood theory. We exemplify our results and method with data from a case-control study of sudden infant death syndrome in which data on some variables were obtained from both maternal interviews and medical records.

Legal blindness among 10-year-old children in Metropolitan Atlanta: prevalence, 1985 to 1987.

The prevalence of legal blindness in 10-year-olds in metropolitan Atlanta was 6.8 per 10,000 during 1985 to 1987. The prevalence was 8.8 per 10,000 in Black boys, 8.6 per 10,000 in White boys, 6.7 per 10,000 in White girls, and 1.8 per 10,000 in Black girls. Retinopathy of prematurity was the most common known cause (1.0 per 10,000). Of the 61 cases, 40 had other disabilities, including 14 with mental retardation, cerebral palsy, and epilepsy. The low prevalence among Black girls and the frequent occurrence of blindness with other disabilities are noteworthy.

Computer-assisted videokeratography of corneal topography after radial keratotomy.

We used computer-assisted videokeratography to compare the topographies of 32 corneas from 23 subjects after radial keratotomy with those of 47 normal corneas from 47 subjects controlled for age and preoperative keratometric and refractive power. Three ophthalmologists independently classified color-coded videokeratographs based on the color-coded pattern of dioptric power distribution and the cross-sectional shape. Corneas that had radial keratotomy exhibited a polygonal pattern not seen in normal eyes; this occurred in 59% of corneas. All normal corneas demonstrated a cross-sectional shape configuration that was steeper centrally than peripherally; 79% of corneas after radial keratotomy had a shape that was flatter centrally than peripherally. After radial keratotomy, the dioptric power increased from the center to the periphery (radius of approximately 4.6 mm) by 2.8 +/- 2.2 diopters (mean +/- SD), with a sharp inflection zone ("paracentral knee") 2.7 mm from the center; normal corneas showed a smooth decrease in power from the center to the periphery of 1.9 +/- 0.5 diopters.

The impact of differential recall on the results of case-control studies.

Many authors have expressed concern that case-control differences in the accuracy of information reported in interviews may create spurious associations in epidemiological investigations. Nevertheless, the impact of differential misclassification on observed associations has not been systematically examined. This paper presents algebraic and graphical analyses of the effect of case-control differences in reporting accuracy on estimates of association and on test size. These analyses suggest that under certain circumstances, even large differences in accuracy may have a minor impact on the results of a study. Study results may be particularly resistant to differences in the sensitivity of recall when the prevalence of exposure is low. The results also illustrate how researchers may evaluate the potential impact of differential misclassification on the validity of their own investigations.

Recall bias in a case-control study of sudden infant death syndrome.

Case-control differences in the accuracy of maternal recall may create spurious associations between suspected risk factors and perinatal conditions. We examined case-control differences in the accuracy of maternal recall and evaluated the impact of maternal reporting errors on observed measures of association. We compared interview information with information recorded on medical records for the mothers of 226 cases of Sudden Infant Death Syndrome (SIDS) and the mothers of 226 living controls. We found that having a child die from SIDS increased the sensitivity of recall for less than half of the 25 study variables. However, for 18 of the 25 variables, the mothers of SIDS cases were more likely than the mothers of living controls to report events that could not be confirmed on medical records. Case-control differences in recall accuracy did not appear to create spurious associations with SIDS or to bias most associations away from the null value.