Search for ethnic, geographic, and other factors in the epidemiology of Down syndrome in South America: analysis of data from the ECLAMC project, 1967-1997.

We have analyzed data on 3,157 cases of Down syndrome (DS) from nine South American countries in consecutive series of hospital live births over a 30-year period, with particular emphasis on possible ethnic or geographic variations in maternal age-adjusted incidence. The data constitute the largest series of DS cases assembled to date from an area lacking advanced health care systems. Absolute incidence rates were estimated from total hospital live births; relative rates were estimated from matched case-control data using conditional logistic regression. Maternal age-adjusted rates were closely similar to those reported elsewhere, and showed little or no dependency on other factors investigated, including paternal age, birth order, ancestral origin, country of birth, maternal educational level, maternal ABO and Rhesus blood groups, interval to and outcome of mother's previous pregnancy, and parental consanguinity. The absence of an effect of high birth order was particularly notable because of the relatively large number of grand multipara resulting from high fertility in this population. The study adds to a body of evidence suggesting that maternal age-adjusted DS rates vary little across human populations, and are therefore unlikely to be greatly influenced by genetic or environmental factors that differ between them. An unusual finding was of a markedly lower sex ratio (98 males per 100 females) than has been reported in other DS samples.

Accuracy of alternative approaches to capture-recapture estimates of disease frequency: internal validity analysis of data from five sources.

The authors used "internal validity analysis" to evaluate the performance of various capture-recapture methods. Data from studies with five overlapping, incomplete lists generated subgroups whose known sizes were compared with estimates derived from various four-source capture-recapture analyses. In 15 data sets unanalyzed previously (five subgroups of each of three new studies), the authors observed a trend toward mean underestimation of the known population size by 16-25%. (Coverage of the 90% confidence intervals associated with the method found to be optimal was acceptable (13/15), despite the downward bias.) The authors conjectured that (with the obvious exception of geographically disparate lists) most data sets used by epidemiologists tend to have a net positive dependence; that is, cases captured by one source are more likely to be captured by some other available source than are cases selected randomly from the population, and this trend results in a bias toward underestimation. Attempts to ensure that the underlying assumptions of the methods are met, such as minimizing (or adjusting adequately) for the possibility of loss due to death or migration, as was undertaken in one exceptional study, appear likely to improve the behavior of these methods.

Recommendations for presentation and evaluation of capture-recapture estimates in epidemiology.

We propose 15 recommendations for approaches to capture-recapture analysis in epidemiology. We apply them to a report of such an analysis of a measles epidemic [McGilchrist et al., J Clinical Epidemiol 1996, 49: 293-296] and to comments thereon by R. C. Cormack [J Clinical Epidemiol 1999; 52: 909-914]. The latter challenged the utility of the data on the measles outbreak for any reliable capture-recapture estimates. We suggest that, adopting the perspective of W. Edwards Deming, one can only make judgments as to the reliability of capture-recapture data, methods, and derived estimates in the light of (i.e., conditional upon) their eventual intended use. Capture-recapture approaches "unreliable" from one perspective may be "reliable," and/or more appropriately, "useful" from another. We consider the utility of ancillary and ad hoc information that may be available or worth seeking to supplement a capture-recapture analysis. We use information within the study of McGilchrist et al. to illustrate how, with such ancillary information, one may overcome the main thrust of the objections of Cormack in situations in which one observes apparently anomalous or hard to understand data structures. Making certain simple assumptions we regard as plausible, we estimate the number of affected in the measles epidemic as between about 700-1300. We derive this from data on 502 cases in a Register, an ad hoc sample of 91 cases in one age group in the general population, and the report of 41 cases in both of these. Our result is only 15-30% the total implied by the estimates McGilchrist et al. derived with more complex methods and many assumptions in addition to our own. We discuss various approaches to evaluating "reliability" of our estimate conditional upon intended uses by policy makers.

International variation in reported livebirth prevalence rates of Down syndrome, adjusted for maternal age.

Reported livebirth prevalence of Down syndrome (DS) may be affected by the maternal age distribution of the population, completeness of ascertainment, accuracy of diagnosis, extent of selective prenatal termination of affected pregnancies, and as yet unidentified genetic and environmental factors. To search for evidence of the latter, we reviewed all published reports in which it was possible to adjust both for effects of maternal age and for selective termination (where relevant). We constructed indices that allowed direct comparisons of prevalence rates after standardising for maternal age. Reference rates were derived from studies previously identified as having near complete ascertainment. An index value significantly different from 1 may result from random fluctuations, as well as from variations in the factors listed above. We found 49 population groups for which an index could be calculated. Methodological descriptions suggested that low values could often be attributed to under-ascertainment. A possible exception concerned African-American groups, though even among these most acceptable studies were compatible with an index value of 1. As we have reported elsewhere, there was also a suggestive increase in rates among US residents of Mexican or Central American origin. Nevertheless, our results suggest that "real" variation between population groups reported to date probably amounts to no more than +/-25%. However, reliable data in many human populations are lacking including, surprisingly, some jurisdictions with relatively advanced health care systems. We suggest that future reports of DS livebirth prevalence should routinely present data that allow calculation of an index standardised for maternal age and adjusted for elective prenatal terminations.

Female predominance (low sex ratio) in 47,+21 mosaics.

Data from the New York State Chromosome Registry on over 10,000 cases of Down syndrome reported from 1977 to 1996 confirm findings in the England and Wales Cytogenetic Register that in mosaic 46/47,+21 cases of Down syndrome the male/female ratio (as inferred from XY and XY karyotypes respectively) is less than 1.0 as opposed to the ratio in nonmosaic 47,+21 cases in which the ratio is close to 1.2, or in the general background population with ratio of about 1.05. These results may reflect in part differential pairing in 47,+21 somatic cells of X and Y chromosomes with a 21 chromosome and/or in in-utero selection.

Elevated maternal age-specific rates of Down syndrome liveborn offspring of women of Mexican and Central American origin in California.

BACKGROUND: Maternal age-specific rates of Down syndrome livebirths are widely utilized in personal and policy decisions concerning provision of and election of prenatal cytogenetic diagnostic services. The only extensive reference data available are on those of primarily European ancestral origin. In the absence of definitive evidence of any ethnic, racial or environmental influence upon rates (other than those associated with age) these rate schedules have been widely applied to those of all national origins. METHODS: Available material age-specific data on livebirths from intensive studies on those of Hispanic (primarily of Mexican and Central American background) and of other origin in populations in the U.S.A. with likely complete ascertainment were analysed. The numbers observed were compared with (i) those predicted from established published rate schedules in those of primarily European origin, and (ii) with the observations on livebirths of non-Hispanic European origin in the same population as the Hispanic live births. RESULTS: In comparisons with the numbers predicted from published rates, observed numbers of case among Hispanic live births were increased by 19 per cent (SE 0.06) in younger mothers, 23 per cent in older mothers (SE 0.07) and 20 per cent (SE 0.04) in those of all ages. Comparisons with observed rates in those of Hispanic origin with those observed in non-Hispanic births in the same time intervals and populations indicated that the excess rates in Hispanics were not attributable to some local factor increasing rates in all ethnic groups at least among those under 35. CONCLUSIONS: Data on mothers of Mexican and Central American origin residing in the U.S.A. indicate maternal age-specific rates of Down syndrome in live births about 20 per cent greater than those in published rate schedules on Down syndrome, widely used in decisions concerning election or provison of prenatal diagnostic services. The reason for this difference remains unknown.

An exact test for all-way interaction in a 2M contingency table: application to interval capture-recapture estimation of population size.

An exact conditional test for an M-way log-linear interaction in a fully observed 2M contingency table is formulated. From this is derived a procedure for interval estimation of the total count N in a 2M contingency table, one of whose entries is unobserved. This procedure has an immediate application to interval estimation of the size of a closed population from incomplete, overlapping lists of records, as in capture-recapture analysis of epidemiological data. Data on the prevalence of spina bifida in live births in upstate New York in 1969-1974 illustrate this application.

Joint estimation of Down syndrome risk and ascertainment rates: a meta-analysis of nine published data sets.

In this paper we present an analysis of nine data sets in which ascertainment and maternal age risk of Down syndrome are estimated jointly using maximum likelihood. We include data on 4825 Down syndrome cases from nine previously published data sets. These include data from studies carried out before the introduction of prenatal screening and from recent studies involving women who had not received prenatal testing. Our results show that, allowing for under-ascertainment, there is a good degree of consistency between the different data sets. We compare the three- and five-parameter constant plus exponential model with a three-parameter logistic model for maternal age-specific risk. We show that the three-parameter logistic model provides a good fit to the data and compare rates from this model with those derived from published studies of uncertain completeness (Cuckle et al., 1987) and those from data sets believed to be complete (Halliday et al., 1995; Hecht and Hook, 1994, 1996). In general, our results agree closely with those of the latter, but achieve greater precision because of the inclusion of additional data. Our derived rates are considerably higher than those of Cuckle et al. (1987), which are embedded in many computer systems for generating risks.

Marginal versus conditional versus 'structural source' models: a rationale for an alternative to log-linear methods for capture-recapture estimates.

Log-linear models for capture-recapture type data are widely used for estimating sizes of populations. Log-linear methods model conditional interactions between the sources. Often, however, the marginal associations are more appropriate and easier for the practitioner to conceptualize. Analyses here of previously published data on cases of spina bifida in upstate New York are used to show how the assumption that sources are conditionally independent can give biased estimates if in fact the sources are marginally independent. A plausible model for the structural sources of interactions between the sources of information about spina bifida cases is developed which implies marginal independence of two of the sources rather than conditional independence. Estimates of the population total based on marginal independence are derived and give larger estimates of the population total than those derived based upon conditional dependence. When investigators can in fact model the likely underlying relationships of the sources in the population, we suggest considering modelling the potential interdependencies of the sources, which we term 'structural source modeling'.

Validity of methods for model selection, weighting for model uncertainty, and small sample adjustment in capture-recapture estimation.

In log-linear capture-recapture approaches to population size, the method of model selection may have a major effect upon the estimate. In addition, the estimate may also be very sensitive if certain cells are null or very sparse, even with the use of multiple sources. The authors evaluated 1) various approaches to the issue of model uncertainty and 2) a small sample correction for three or more sources recently proposed by Hook and Regal. The authors compared the estimates derived using 1) three different information criteria that included Akaike's Information Criterion (AIC) and two alternative formulations of the Bayesian Information Criterion (BIC), one proposed by Draper ("two pi") and one by Schwarz ("not two pi"); 2) two related methods of weighting estimates associated with models; 3) the independent model; and 4) the saturated model, with the known totals in 20 different populations studied by five separate groups of investigators. For each method, we also compared the estimate derived with or without the proposed small sample correction. At least in these data sets, the use of AIC appeared on balance to be preferable. The BIC formulation suggested by Draper appeared slightly preferable to that suggested by Schwarz. Adjustment for model uncertainty appears to improve results slightly. The proposed small sample correction appeared to diminish relative log bias but only when sparse cells were present. Otherwise, its use tended to increase relative log bias. Use of the saturated model (with or without the small sample correction) appears to be optimal if the associated interval is not uselessly large, and if one can plausibly exclude an all-source interaction. All other approaches led to an estimate that was too low by about one standard deviation.