Additional analyses exploring the interrelationships between birth outcomes and their main and conjoint effects on infant mortality would seem to hold considerable potential for generating insights into variation in perinatal health. More population-specific studies based on deviations from birth outcome survival optimums are needed, as alternatives to or in conjunction with, studies using conventional indicators of birth weight and gestational age.
We also need answers to the question of why the relative disparity between black and white infant mortality is increasing (Guyer et al. 1998). It would seem that part of the explanation likely resides in the erosion of the black survival advantage at short gestations and very low birth weights, which in turn may partially be due to the differential benefits of pulmonary surfactant therapy for neonates with RDS. Clinical study supports such an explanation. However, one recent study suggests that differential access to medical intervention is a likely explanation (Frisbie et al. 2004), One obstacle to large-scale studies is that, although infant deaths in which RDS is implicated can be identified from NCHS linked files, there is no record in these files of whether surfactant therapy was applied. Further, a comparison of the records of neonatal intensive care nursery systems in St. Louis with NCHS linked files indicated substantial underreporting of RDS (Hamvas et al. 1998).
Given the dearth of contextual research in this area, it seems eminently worthwhile to mount an effort to develop data sets designed especially for multilevel analyses. Moreover, using whatever relevant data are available, multilevel studies of pregnancy outcomes might profit from the application of random effects models. In the simplest case, the intercept is allowed to vary across areas (say, neighborhoods). Significant intercept variation ''suggests either that neighborhood-level factors (or alternatively, omitted individual-level factors closely associated with neighborhoods) may be related to average outcomes for neighborhoods'' (Diez-Roux 2000:181). Random slope models would allow investigation of whether individual-level effects vary by area (Diez-Roux 2000: 182).
For the demography of health there is perhaps no more pressing need than to collaborate with social epidemiologists and medical researchers in the quest to identify biological mechanisms. In general, without such an effort, social scientists ''are only able to infer physiologic pathways and therefore cannot rule out alternative explanations typically advocated by biomedical researchers'' (Fremont and Bird 1999: 126). The need for interdisciplinary research of this sort is most palpable where the outcome of interest is a biological event, and no biological event should be of any greater interest to us than the survival and health of infants and children.
Was this article helpful?
If Pregnancy Is Something That Frightens You, It's Time To Convert Your Fear Into Joy. Ready To Give Birth To A Child? Is The New Status Hitting Your State Of Mind? Are You Still Scared To Undergo All The Pain That Your Best Friend Underwent Just A Few Days Back? Not Convinced With The Answers Given By The Experts?