No description available.

Uncertainty is recognized as a significant perceptual variable in the experience of illness. The purpose of this study was to gain an understanding of how parents of ill infants cope with the uncertainty inherent in illness-related events and situations. A conceptual framework of coping with uncertainty was proposed and tested. The four constructs in the model and their measures were perceived uncertainty (Mishel's Parents' Perceptions of Uncertainty Scale), cognitive appraisal (Lazarus and Folkman's Appraisal Questionnaire), coping efforts (Lazarus and Folkman's Ways of Coping Checklist) and cognitive schema (Mishel's Parents' Perceptions of Uncertainty Scale and grounded theory methodology). Methodological triangulation was used. A quantitative, longitudinal, descriptive correlational design examined the model. A qualitative study using grounded theory methodology explored the forming and using of a cognitive schema. A convenience sample of 37 parents of critically ill neonates participated in the quantitative study with 15 of those also participating in the qualitative study. Self report questionnaires measured model variables. Interviews comprised the grounded theory approach. Descriptive and correlational statistics characterized model variables and their relationships. Constant comparative analysis identified processes central to forming and using a cognitive schema. From the descriptive results, parents perceived high levels of uncertainty when measured at approximately 2.5 days following the ill infant's birth. Appraisal of uncertainty as harmful to well-bring was correlated with perceived ambiguity (r =.63) and complexity (r =.36). The coping efforts of self-blame (r =.53) and wishful thinking (r =.44) were related to the harm appraisal. Significant decreases in perceived ambiguity and lack of information were demonstrated when uncertainty was measured again at approximately eight days following birth. From the grounded theory methodology, three processes central to cognitive schema were identified (forming, framing, using) were discovered and conceptualized. When uncertainty was perceived, parents actively sought information in forming a schema. With sufficient information, information was categorized to frame an explanation of illness experiences. With framing, schema was created and used by the parents. Methodological triangulation accounted for consistencies and inconsistencies across quantitative and qualitative results. The model of coping with uncertainty was supported through triangulation.

The ultra-processing of food has become a much more important aspect of dietary patterns and dietary quality in terms of its impact on body weight, diet related diseases, health, and well-being in the past decades. NOVA is a set of guidelines developed that classifies diet quality by degree of food processing. The NOVA guidelines distinguish four categories: unprocessed /minimally processed foods; culinary ingredients; processed foods; and ultra-processed foods. Numerous studies have found an association of ultra-processed foods and health conditions such as obesity and metabolic syndrome. This study analyzed the associations between maternal diet quality as measured by NOVA and maternal anthropometric and neonatal body composition outcomes. The optimal method of nutrition intervention and education for this special population remains unknown; using NOVA may provide researchers with a different lens to assess diet quality and health care professionals with additional vocabulary to convey more tailored messages regarding optimal nutrition strategies for mother and offspring. Using data collected from a large randomized controlled intervention trial at pre and post intervention, this study aimed to compare the NOVA guidelines assessment of maternal diet quality to the parent study assessment of diet quality, the Healthy Eating Index (HEI), using statistical correlations. Secondly, this study aimed to look at the relationship of ultra-processed food intake to the maternal gestational weight gain experience using a logistic regression. Thirdly, this dissertation aimed to explore the relationship between maternal ultra-processed food intake and neonatal lean mass as measured by quantitative magnetic resonance (QMR) and fat free mass as measured by air displacement plethysmography (ADP: PEAPOD). In terms of maternal outcomes, the study found that NOVA and HEI were significantly correlated at pre intervention but not at post intervention. The odds of gaining excessive gestational weight decreased as maternal ultra-processed food intake increased - which was not in the hypothesized direction - when using study participant data. However, the odds of gaining excessive gestational weight increased as maternal ultra-processed food intake increased - which was in the hypothesized direction - when using the Institute of Medicine weight gain recommendations. Also, while obesity did not predict excessive gestational weight gain, those with obesity ultra-processed food intake did predict gestational weight gain. These various inconsistencies are likely due to the instability of the dietary intake data because only one 24 -hour dietary recall was obtained from mother. In addition, the mothers’ diets were very healthy to begin with, where ultra-processed food intake formed about 45% of calories both pre and post intervention, when the national average is 57%. Race was also significant predictors of gestational weight gain for the mothers. Being non-white significantly increased the odds of gaining excessively as did the interaction of having obesity and eating more ultra-processed foods. In terms of neonatal outcomes, findings from this study suggest that length and fat mass are significant predictors of lean mass in neonates. In terms of the impact of maternal ultra-processed food intake, the higher the consumption of ultra-processed food, the greater the neonatal lean mass, which this was not in the hypothesized direction. However, the association was minimal with very small beta weights and regression line, when plotted was quite flat, so that the finding is not clinically meaningful. It remains important to know whether maternal ultra-processed food intake influences gestational weight gain and the body composition of the neonate. Thus, future research should include using similar data analyses on a population with a more nationally representative diet, a larger sample size, and a more robust measure of dietary intake such as three 24-hour recalls. Given that a similar recent study found ultra-processed food to be highly predictive of maternal and neonatal outcomes, and many other studies have demonstrated that ultra-processed food is related to several health conditions in many countries that this study did not measure, it seems prudent for healthcare providers to take advantage of prenatal visits as a window of opportunity to encourage the consumption of unprocessed and minimally foods and help women make informed decisions regarding ultra-processed foods.

The Hispanic birthweight paradox, whereby Hispanic women exhibit a comparable or lower risk of bearing a low birthweight infant than their white counterparts despite relative socioeconomic disadvantage, has been observed across a number of research studies. However, the majority of evidence for the paradox has focused on Hispanics in aggregate form or on populations with primarily Mexican ancestry and has relied largely on outcome measures with important methodological shortcomings. Furthermore, studies have identified the variation of birthweight risk among Hispanics by nativity, maternal education and neighborhood composition, but the evidence has been scarce and inconsistent. The overall goal of this dissertation was to investigate the Hispanic health paradox with relation to measures of birthweight and infant size in births to women residing in New York City aged 20 years and older, using birth records for years 2003 through 2007 collected by the Office of Vital Statistics of the New York City Department of Health and Mental Hygiene (N=460,881). The main outcomes of interest in this study were mean birthweight, low birthweight (LBW, defined as < 2500 grams versus ≥ 2500 grams) and small for gestational age (SGA, calculated as the 10th percentile for birthweight at each week of gestational age and by sex). Multilevel logistic models with random effects were used to estimate odds ratios for the association between race/ethnicity and measures of birthweight and infant size, while controlling for individual-level and contextual factors and accounting for the correlation between observations within the same neighborhood. Analyses were conducted with Hispanics as an aggregate group as well as with race/ethnic-nativity subgroups. In addition, effect measure modification by maternal education and by neighborhood proportion of Hispanic population (NPHP) were examined. This research confirmed the Hispanic paradox in SGA analyses for Hispanics overall and for both U.S.-born and foreign-born Hispanics, but not in analyses with LBW or with mean birthweight. As compared to white women, black women exhibited 50% greater risk (OR:1.50;95%CI:1.45,1.55) and Hispanic women comparable risk (OR:1.03;95%CI:1.00,1.06) of having an SGA infant, in a fully adjusted model. With regard to LBW, the risk was more than double for black women (OR:2.25;95%CI:2.16,2.35) and close to 50% greater for Hispanic women (OR:1.46;95%CI:1.40,1.53) as compared to that of their white counterparts. In addition, the mean birthweight of infants born to Hispanic women was significantly lower compared to those born to white women. Furthermore, the relationship between race/ethnicity and all three measures of birthweight and infant size varied by maternal nativity status (p<0.0001), with infants of foreign-born women experiencing more favorable outcomes relative to their U.S.-born counterparts. The paradox with SGA was also apparent across most Hispanic race/ethnicity-nativity subgroups, The odds were greatest among black and Puerto Rican women overall (OR:1.52;95%CI:1.47,1.57 and OR:1.17;95%CI:1.13,1.22, respectively) and lowest among Mexican and South American women overall, (OR:0.91;95%CI:0.87,0.95 and OR:0.85;95%CI:0.80,0.89) as compared to white women in a fully adjusted model. The odds of SGA for infants born to Dominicans, Central Americans and Cubans in the fully adjusted model were similar to those born to whites. In addition, SGA varied by maternal nativity status (p<0.0001), with more favorable SGA odds observed among infants of most foreign-born women, as compared to whites. The exception was U.S.-born Puerto Ricans who consistently exhibited elevated risk of SGA relative to whites. The association of race/ethnicity-nativity with SGA varied by maternal educational attainment (p<0.0001), but the influence varied by subgroup. The observed advantage of foreign birth was stronger among less educated women of all Hispanic subgroups other than Puerto Ricans and Cubans. Similarly, the variation of SGA risk by neighborhood proportion of Hispanic population (NPHP) differed across subgroups (p<0.0001). NPHP did not appear to influence the association between race/ethnicity-nativity and SGA in a consistent pattern, but among black women and US-born Puerto Rican women greater NPHP was associated with a higher risk of SGA. Findings from this study underscore the importance of using SGA an accurate measure of infant size and of conducting analyses disaggregating race/ethnicity and nativity subgroups. Future research should focus on factors that contribute to the resilience of Hispanic subgroups in the face of adverse economic circumstances, such as the role of social support networks and acculturation. Greater understanding of the salubrious circumstances that lower the risk of adverse birth outcomes has major public health benefits, especially for a wide-ranging population of mothers, Hispanic and non-Hispanic, and their infants.

No description available.

This dissertation consists of three essays in health economics, paying special attention to neonatal care provision and newborn health outcomes in the United States. The first chapter evaluates physician productivity, focusing on the matching between physician skills and patient conditions. High U.S. spending on health care is commonly attributed to its intensity of specialized, high-tech medical care. A growing body of research focuses on physicians whose medical decisions shape treatment intensity, costs, and patient outcomes. Often overlooked in this research is the assignment of physician skills to patient conditions, which may strongly affect health outcomes and productivity. This matching may be especially important in the case of hospital admissions as high-frequency fluctuations in patient flow make it challenging to maintain effective matches between the best-suited physicians and their patients. This paper focuses on hospitals’ responses to demand shocks induced by unscheduled high-risk admissions. I show that these demand shocks result in physician–patient mismatches when hospitals are congested. Specifically, highly specialized physicians who are brought in to treat unscheduled high-risk admissions also treat previously admitted lower-risk patients. This leads to increased treatment intensity for lower-risk patients, which I attribute to persistence in physician practice style. Despite the greater treatment intensity, I find no detectable improvement in health outcomes, which prima facie could be viewed as waste. However, the mismatches observed only at high congestion levels more likely reflect hospitals’ careful assessment of costs and benefits when assigning physicians to patients – maintaining preferred physician–patient matching can be particularly costly when congestion is high. My findings highlight the need to consider both heterogeneity within patient and physician type, and furthermore show how the common phenomenon of demand uncertainty can promote mismatch between these types. The second chapter assesses hospital self-reported facility data quality using annual Institutional Cost Report (ICR). In the United States, hospital facilities are under public and government supervision. The central motivation behind this is that overbuilding and redundancy in health care facilities will lead to overutilization and higher health care costs. However, little is known about the effectiveness of these facility regulation policies. Taking certified capacities recorded by the Department of Health as reliable benchmarks, this paper presents evidence that hospitals upcode their neonatal intensive care unit (NICU) bed levels when reporting capacities in ICR. Reported NICU utilization in ICR is mostly under the top level NICU bed, which matches the bed capacity upcoding pattern. This indicates either significant overutilization which leads to NICU overcrowding, or upcoding in medical billing that results in inflated medical charges. Findings in this paper point to a potentially effective way for regulators and insurers to limit overutilization – improving hospitals’ compliance with their certified capacities. This paper also provides important guidelines for a large body of research that uses ICR data by developing an assessment of ICR data quality. The third chapter, which is joint work with Douglas Almond, measures gender inequality in perinatal health among Chinese-American newborns. The literature on “missing girls" suggests a net preference for sons both in China and among Chinese immigrants to the West. Perhaps surprisingly, we find that newborn Chinese-American girls are treated more intensively in US hospitals: they are kept longer following delivery, have more medical procedures performed, and have more hospital charges than predicted (by the non-Chinese gender difference). What might explain more aggressive medical treatment? We posit that hospitals are responding to worse health at birth of Chinese-American girls. We document higher rates of low birth weight, congenital anomalies, maternal hypertension, and lower APGAR scores among Chinese American girls – outcomes recorded prior to intensive neonatal medical care and relative to the non-Chinese gender gap. To the best of our knowledge, we are the first to find that son preference may also compromise “survivor" health at birth. On net, compromised newborn health seems to outweigh the benefit of more aggressive neonatal hospital care for girls. Relative to non-Chinese gender differences, death on the first day of life and in the post-neonatal period is more common among Chinese-American girls, i.e. later than sex selection is typically believed to occur.

published_or_final_version / Community Medicine / Doctoral / Doctor of Philosophy

published_or_final_version / Medicine / Master / Doctor of Medicine

Background: Statistical and machine learning techniques are now being incorporated into high-dimensional mixture research to overcome issues with traditional methods. Though some methods perform well on specific tasks, no method consistently outperforms all others in complex mixture analyses, largely because different methods were developed to answer different research questions. The research presented here concentrates on answering a single mixtures question: Are there exposure patterns within a mixture corresponding with sources or behaviors that give rise to exposure? Objective: This dissertation details work to design, adapt, and apply pattern recognition methods to environmental mixtures and introduces two methods adapted to specific challenges of environmental health data, (1) Principal Component Pursuit (PCP) and (2) Bayesian non-parametric non-negative matrix factorization (BN²MF). We build on this work to characterize the relationship between identified patterns of in utero endocrine disrupting chemical (EDC) exposure and child neurodevelopment. Methods: PCP---a dimensionality reduction technique in computer vision---decomposes the exposure mixture into a low-rank matrix of consistent patterns and a sparse matrix of unique or extreme exposure events. We incorporated two existing PCP extensions that suit environmental data, (1) a non-convex rank penalty, and (2) a formulation that removes the need for parameter tuning. We further adapted PCP to accommodate environmental mixtures by including (1) a non-negativity constraint, (2) a modified algorithm to allow for missing values, and (3) a separate penalty for measurements below the limit of detection (PCP-LOD). BN²MF decomposes the exposure mixture into three parts, (1) a matrix of chemical loadings on identified patterns, (2) a matrix of individual scores on identified patterns, and (3) and diagonal matrix of pattern weights. It places non-negative continuous priors on pattern loadings, weights, and individual scores and uses a non-parametric sparse prior on the pattern weights to estimate the optimal number. We extended BN²MF to explicitly account for uncertainty in identified patterns by estimating the full distribution of scores and loadings. To test both methods, we simulated data to represent environmental mixtures with various structures, altering the level of complexity in the patterns, the noise level, the number of patterns, the size of the mixture, and the sample size. We evaluated PCP-LOD's performance against principal component analysis (PCA), and we evaluated BN²MF's performance against PCA, factor analysis, and frequentist nonnegative matrix factorization (NMF). For all methods, we compared their solutions with true simulated values to measure performance. We further assessed BN²MF's coverage of true simulated scores. We applied PCP-LOD to an exposure mixture of 21 persistent organic pollutants (POPs) measured in 1,000 U.S. adults from the 2001--2002 National Health and Nutrition Examination Survey (NHANES). We applied BN²MF to an exposure mixture of 17 EDCs measured in 343 pregnant women in the Columbia Center for Children’s Environmental Health's Mothers and Newborns Cohort. Finally, we designed a two-stage Bayesian hierarchical model to estimate health effects of environmental exposure patterns while incorporating the uncertainty of pattern identification. In the first stage, we identified EDC exposure patterns using BN²MF. In the second stage, we included individual pattern scores and their distributions as exposures of interest in a hierarchical regression model, with child IQ as the outcome, adjusting for potential confounders. We present sex-specific results. Results: PCP-LOD recovered the true number of patterns through cross-validation for all simulations; based on an a priori specified criterion, PCA recovered the true number of patterns in 32% of simulations. PCP-LOD achieved lower relative predictive error than PCA for all simulated datasets with up to 50% of the data < LOD. When 75% of values were < LOD, PCP-LOD outperformed PCA only when noise was low. In the POP mixture, PCP-LOD identified a rank three underlying structure. One pattern represented comprehensive exposure to all POPs. The other two patterns grouped chemicals based on known properties such as structure and toxicity. PCP-LOD also separated 6% of values as extreme events. Most participants had no extreme exposures (44%) or only extremely low exposures (18%). BN²MF estimated the true number of patterns for 99% of simulated datasets. BN²MF's variational confidence intervals achieved 95% coverage across all levels of structural complexity with up to 40% added noise. BN²MF performed comparably with frequentist methods in terms of overall prediction and estimation of underlying loadings and scores. We identified two patterns of EDC exposure in pregnant women, corresponding with diet and personal care product use as potentially separate sources or behaviors leading to exposure. The diet pattern expressed exposure to phthalates and BPA. One standard deviation increase in this pattern was associated with a decrease of 3.5 IQ points (95% credible interval: -6.7, -0.3), on average, in female children but not in males. The personal care product pattern represented exposure to phenols, including parabens, and diethyl phthalate. We found no associations between this pattern and child cognition. Conclusion: PCP-LOD and BN^2MF address limitations of existing pattern recognition methods employed in this field such as user-specified pattern number, lack of interpretability of patterns in terms of human understanding, influence of outlying values, and lack of uncertainty quantification. Both methods identified patterns that grouped chemicals based on known sources (e.g., diet), behaviors (e.g., personal care product use), or properties (e.g., structure and toxicity). Phthalates and BPA found in food packaging and can linings formed a BN²MF-identified pattern of EDC exposure negatively associated with female child intelligence in the Mothers and Newborns cohort. Results may be used to inform interventions designed to target modifiable behavior or regulations to act on dietary exposure sources.

Reducing health disparities and achieving health equity in maternal and infant health is a critical concern for social work and public health stakeholders more generally. This three-paper dissertation is dedicated to exploring program or policy modifiable social determinants of maternal and infant health with a particular focus on vulnerable populations. Paper one explores the influence of household members on women’s sexual and reproductive behaviors. Paper two studies the impact of smoke-free regulations on birth outcomes in Latin America. Lastly, paper three looks at the effects of police use of force, and racialized police use of force, on maternal and infant health. The findings of these papers provide important information to inform programs and policies aimed at improving reproductive health and well-being in the U.S. and Latin America.