Trisomy 21 (T21), trisomy 18 (T18) and trisomy 13 (T13) represent the most common autosomal trisomies detected in live-born infants. Previous studies have addressed interventions, morbidities and survival in term or near-term infants with T21, T18 or T13, or were limited by a small number of patients. However, the combination of one of these chromosomal anomalies and very low birth weight (VLBW) presents greater challenges. Data from the NICHD Neonatal Research Network (NRN) and from the Vermont Oxford Network (VON) databases were used to examine the frequency, interventions, risk of mortality and neonatal morbidities, including patent ductus arteriosus (PDA), necrotizing enterocolitis (NEC), late onset sepsis (LOS), retinopathy of prematurity (ROP), and bronchopulmonary dysplasia (BPD), among VLBW infants with T21, T18 or T13 compared to VLBW infants without major birth defects (BD) and VLBW infants with non-chromosomal BD. Anthropometric VON charts for the assessment of birth weight for gestational age among 22 week to term infants with T21, T18 or T13 were also developed. In the VON database (n=539,509), the frequency of VLBW infants diagnosed with T21 was 1681 (0.31%), with T18 was 1416 (0.27%), and with T13 was 435 (0.08%). Major surgery was reported for 30.4% of infants with T21, 9.2% with T18, and 6.8% with T13. In-hospital mortality occurred for 33.1% of infants with T21, 89.0% with T18, and 92.4% with T13. Median survival time was 4 days (95% CI: 3-4) among infants with T18 and 3 days (95% CI: 2-4) among infants with T13. Birth weight for gestational age charts were created using VON data with a total of 5147 infants with T21 aged 22-41 weeks, 1053 infants with T18 aged 22-41 weeks, and 613 infants with T13 aged 22-40 weeks. Among the three groups, infants with T18 were the most likely to be growth restricted while infants with T21 were the least likely to be growth restricted. The new anthropometric VON charts for infants with T21 were also compared to the Lubchenco and Fenton charts and both showed frequent misclassification of infants with T21 as small or large for gestational age. In the NICHD NRN database (n=52,259), 133 (0.26%) VLBW infants were diagnosed with T21, 132 (0.25%) with T18 and 40 (0.08%) with T13. The adjusted relative risk, estimated using Poisson regression models with robust variance estimators, showed an increased risk of death, PDA, NEC, LOS, and BPD among infants with T21 relative to infants with no BD. Relative to infants with non-chromosomal BD, infants with T21 were at increased risk of PDA and NEC. A trend toward a lower risk of ROP was observed among infants with T21 compared to infants with non-chromosomal BD and infants without major BD. Infants with T13, but not infants with T18, were less likely to be mechanically ventilated than infants with T21 and infants without BD. Infants with T18 had increased risk of PDA compared to infants with T13, infants with T21 and infants without BD and increased risk of BPD compared to infants with T21 and infants without BD. The current studies evaluated the largest cohorts of VLBW infants with T21, T18 or T13. These data are important to help families and care providers make informed decisions involving the care of their VLBW infants with these chromosomal anomalies.

The goal of this dissertation research was to better understand relationships among physical activity (PA), cardiorespiratory fitness, adiposity, and cardiovascular (CV) health in children and adolescents. The aim of the first paper was to examine whether fitness and adiposity are independently associated with CV risk factors during puberty. Study participants were 126 prepubertal Caucasian children participating in a longitudinal four-year follow-up study. Fitness level was determined by VO2 max (L/min) obtained from maximal graded exercise testing and adiposity level was determined by the sum of skinfolds. Gender-specific individual growth curve models, including both VO2 max and the sum of skinfolds simultaneously, were fit to predict CV risk factor variables. Models also included covariates such as age, height, weight, and pubertal stage by the Tanner criteria. In both boys and girls, total cholesterol, triglyceride, LDL-C, and systolic blood pressure percentile were positively associated with the sum of skinfolds (P < 0.05), but not with VO2 max (P > 0.05). In conclusion, fitness was not associated with CV risk factors, after adjusting for adiposity, among healthy adolescents. This study suggests that adiposity may play a role in the mechanism underlying the effect of fitness on CV health during puberty. The aim of the second paper was to examine whether early adiposity level is inversely associated with subsequent PA behaviors in childhood. Study participants were 326 children participating in the Iowa Bone Development Study. PA and fat mass were measured using accelerometers and dual energy X-ray absorptiometry (DXA) at approximately 5, 8, and 11 years of age. Gender-specific generalized linear models were fit to examine the association between percent body fat (BF%) at age 8 and intensity-weighted moderate- to vigorous-intensity PA (IW-MVPA) at age 11. After adjusting for IW-MVPA at age 8, an interval between the age 5 and 8 examinations, residualized change scores of BF% and IW-MVPA from age 5 to 8 and mother's education level, BF% at age 8 was inversely associated with IW-MVPA at age 11 among boys (P < 0.05). After adjusting for IW-MVPA at age 8, physical maturity, and family income, BF% at age 8 was inversely associated with IW-MVPA at age 11 among girls (P < 0.05). Categorical analysis also showed that the odd of being in the lowest quartile relative to the highest quartile of IW-MVPA at three-year follow-up for boys and girls with high BF% was approximately four times higher than the odd for those with low BF% (P < 0.05). This study suggests that adiposity levels may be a determinant of PA behavior. Specific intervention strategies for overweight children may be needed to promote PA. The aim of the third paper was to examine whether accelerometer-measured daily light-intensity PA is inversely associated with DXA-derived body fat mass during childhood. The study sample was 577 children participating in the longitudinal Iowa Bone Development Study. Fat mass and PA were measured at about 5, 8, and 11 years of age. Two PA indicators were used, applying two accelerometer count cut-points: the daily sum of accelerometer counts during light-intensity PA (IW-LPA) and the daily sums of accelerometer counts during high-light-intensity PA (IW-HLPA). Measurement time point- and gender-specific multivariable linear regression models were fit to predict fat mass based on IW-LPA and IW-HLPA, including covariates, such as age, birth weight, fat-free mass, height, IW-MVPA and maturity (only for girls). Among boys, both IW-LPA and IW-HLPA were inversely associated with fat mass at age 11 (P < 0.05), but not at ages 5 and 8. Among girls, both LPA variables were inversely associated with fat mass at ages 8 and 11 (P < 0.10 for LPA at age 11, P < 0.05 for others), but not at age 5. In conclusion, this study suggests that light-intensity PA may have a preventive effect against adiposity among older children.

The research presented in this dissertation harnesses two secondary data sources, administrative databases of patient-level healthcare data and adverse event (AE) data reported in randomized clinical trials (RCTs), to assess the relationship between intravenous immune globulin (IVIg) and the risk of clinically serious thromboembolic adverse events (TEEs). Since 2013, IVIg products have carried a boxed warning concerning TEE risk, a determination supported by numerous case reports, a large claims-based risk assessment, and laboratory evaluations of the thrombogenecity of IVIg products. Questions remain concerning the magnitude of the risk overall and across subgroups of IVIg users. Taken together, our results are compatible with the conclusion that the absolute risk of TEE following IVIg use is likely to be low overall. While these results are reassuring, a clinically meaningful elevation in risk cannot be ruled out in certain patient sub-groups, such as older adults and others with a high baseline risk of TEE. A limitation of our research is that differences in TEE risk across products could not be evaluated with sufficient statistical power.

Intimate partner violence (IPV) against women is a major public health issue worldwide. The purpose of this dissertation is to characterize violence perpetrated by an intimate partner against a "high-risk" group of pregnant women who sought elective abortion services at a family planning clinic. Analyses were based on the Iowa Women's Health Experience Survey (IWHES), a cross-sectional study of 519 abortion patients who completed an anonymous, self-administered questionnaire over a period of seven months. IWHES eligibility criteria were 'Seeking pregnancy termination'; 'Age ≥ 18 years'; 'Iowa resident' and 'Fluent in English or Spanish'. The survey instrument covered physical, sexual and psychological types of violence, health correlates of violence as well as demographic, socioeconomic and lifestyle characteristics of participating women and their current intimate partners. Aim I examined the prevalence of physical, sexual and/or psychological abuse by employment characteristics of elective abortion patients and their current intimate partners. Aim II examined associations of substance use, depression and social support with physical, sexual and/or psychological abuse perpetrated by current intimate partners against women seeking pregnancy termination. To achieve the analytic goals of Aims I and II, the study sample was restricted to women who had a current partner and valid IPV data. The overall prevalence of physical, sexual and/or psychological abuse perpetrated by a current partner was 12.3%, with some overlap between the different IPV sub-types. In general, the prevalence of IPV did not differ significantly by employment status or by broadly defined occupational groups of women and their partners. However, a trend was noted whereby a woman's employment and a partner's unemployment were associated with greater likelihood of IPV. Specifically, the prevalence of IPV was highest among couples where the woman was employed and the partner was unemployed. Consistently positive associations were noted between the partner's (but not the woman's) substance use indicators (alcohol intake, binge drinking, recreational drug use) and IPV. Higher levels of depressive symptoms and less perceived availability of social support were noted among women who had experienced IPV versus those who had not experienced IPV. The association between depressive symptoms and IPV was stronger for women who reported having children in their homes compared to those did not report having children in their homes. Implications for policy and future research are discussed.

As skin cancer rates increase, it has become more important for at risk individuals to reduce ultraviolet radiation (UVR) exposure. Limited information is available on characteristics and behaviors of sunless tanning product users in populations with high sun-seeking behaviors. This information is important because use of sunless tanning products could reduce tanning through UVR exposure thereby leading to a reduction in skin cancer. Sorority and fraternity students (n=163) completed a self-administered questionnaire examining sun exposure and tanning attitudes, behaviors and beliefs. Characteristics of sunless tanning product users were compared to non-users using logistic regression accounting for potential clustering effects within sororities and fraternities. Among students surveyed, 34% reported ever using sunless tanning products. Ever users of sunless tanning products were significantly more likely to be female (OR=7.5), have fair skin (OR=1.4), have used tanning beds greater than 50 times (OR=2.5), and reapply sunscreen when outside on a sunny day (OR=1.3). Ever users of sunless tanning products and those with a preference for these products because they are safer than tanning beds or sunbathing were more likely to reapply sunscreen on a sunny day in the summer. However, other sun protection behaviors (i.e. sunscreen use, amount of sunscreen used, and avoidance of midday sun while on spring break) were not more likely to be adopted by these students. Prevention efforts could target these fair-skinned females to increase their use of sunless tanning products in combination with sunscreen use and reapplication along with avoidance of midday sun.

A pediatric intensivist in the University of Iowa Hospitals and Clinic's (UIHC) Pediatric Intensive Care Unit (PICU) was concerned about antimicrobial use in the unit. However, no one had quantified antimicrobial use in the UIHC's PICU or described the patterns of antimicrobial use in this unit. To address the intensivist's concern, the principal investigator (PI) conducted a retrospective study to determine the percentage of patients who received antimicrobial treatments, to determine the indications for antimicrobial use, and to identify antimicrobial agents used most frequently in the unit. On basis of our data, we hypothesized that empiric antimicrobial use, particularly the duration of therapy, could be decreased. We implemented a six-month intervention during which we asked the pediatric intensivists to complete an antimicrobial assessment form (AA) to document their rationale for starting antimicrobial treatments. We postulated that this documentation process might remind physicians to review antimicrobial therapies, especially empiric therapies, when the microbiologic data became available. In addition, we utilized the AA form to identify factors pediatric intensivists considered when deciding to prescribe empiric antimicrobial treatments. Data from the AA forms suggested that pediatric intensivists in the UIHC's PICU often considered elevated C-reactive protein, elevated white blood cell counts, and elevated temperatures when deciding to start empiric antimicrobial therapy. Data from the three nested periods showed that the median duration of empiric and targeted treatments decreased during the intervention and remained stable during the post-intervention period. The PI estimated that 193 days of empiric antimicrobial therapy and 59 days of targeted antimicrobial therapy, respectively, may have been saved by the decreased durations of therapy. Time series analysis assessing the trend in use of piperacillin-tazobactam, cefepime, and ceftriaxone (measured in mg/wk) did not reveal a significant change over time. On the basis of our results, an intervention strategy using an AA form alone may not be an effective strategy for antimicrobial stewardship in PICUs. Additional measures such as automatic stop orders and computer decision support may be useful for reducing the duration of empiric therapy in PICUs.

The acute nervous system toxicity of organophosphate (OP) pesticides is well described. However, the reported long-term effects of OP pesticides on the nervous system are inconsistent. This inconsistency may be due to imprecise estimates of pesticide exposure, variability of central nervous system (CNS) and peripheral nervous system (PNS) assessment, small samples, and poor control of confounding. The primary goal of this research was to examine the association between long-term OP pesticide use on CNS and PNS function among pesticide applicators. An additional goal was to examine the association between high pesticide exposure events (HPEEs), which typically do not result in acute toxicity, and CNS function. Study participants were recruited from among applicators enrolled in the Agricultural Health Study (AHS) in Iowa and North Carolina. In 2006-2008, 701 male pesticide applicators completed a battery of neurobehavioral (NB) and neurological tests. Information about individual pesticide use was obtained from previous AHS interviews and a questionnaire administered during NB testing. Associations between pesticide use and neurological outcomes were estimated with linear and logistic regression models while controlling for covariates. When associations were examined between agent-specific pesticide use and nine NB tests, significantly poorer performance was observed on four tests and significantly better performance on five tests. Additionally, for some pesticides, we observed differential associations by state, suggesting that regional differences in pesticide practices may influence neurotoxicity. Overall, our results did not provide strong evidence that OP pesticide use was associated with adverse NB test performance. A history of at least one HPEE was reported by 23 percent of participants. Significant adverse associations were observed between HPEEs and two of the nine NB tests. Participants with HPEEs were, on average, 4.9 seconds slower on a test of visual scanning/processing, and 2.2 seconds slower on a test of visual scanning/motor speed. Overall, small but meaningful associations were observed between HPEEs and adverse CNS function. When associations were examined between pesticide use and PNS function, five of six neurological physical examination outcomes were associated with ever-use of one or more OP pesticides. Odds ratios ranged from 1.9 to 3.1. However, mostly null associations were observed between OP pesticide use and electrophysiological tests, hand strength, sway speed and vibrotactile threshold. This study provides some evidence that long-term exposure to OP pesticides is associated with impaired PNS function. In summary, our results suggest that exposure to a few individual OP pesticides as well as HPEEs may contribute to adverse neurological function. The observed exposure-effect associations were present after adjustment for confounding and were independent of past-diagnosed pesticide poisoning. We believe this research contributes important new evidence to an inconsistent literature. Reducing pesticide exposure and preventing HPEEs among pesticide applicators remain important public health goals.

Angiotensin Converting Enzyme (ACE) inhibitors are one of the recommended prevention therapy for patients with acute myocardial infarction (AMI) in the clinical guidelines. Two types (tissue and non-tissue) of ACE inhibitors are available with huge cost difference but the comparative treatment benefit and risk between them are unclear. The objective of this study was to investigate the comparative treatment effectiveness and safety between tissue and non-tissue ACE inhibitors among elderly patients after AMI. This is a retrospective cohort study with intention to treatment design using Medicare service claims files from 2007 to 2009 with Medicare beneficiaries 65 years or older after the index AMI hospitalization and who survived to discharge between January 1 2008 to December 31 2008 and received ACE inhibitors (N=34,679). Risk adjustment and instrumental variable (IV) analyses were used to investigate comparative treatment effectiveness including AMI, stroke, heart failure requiring hospitalization, all-cause mortality and a composite of the endpoints during the follow-up and the comparative treatment safety - a composite of hyperkalemia and acute renal failure requiring hospitalization during the follow-up. Both the risk adjustment and IV analyses showed no significant differences between tissue and non-tissue ACE inhibitors for the investigated outcomes of the comparative treatment effectiveness and safety in the study cohort. However, subgroup analyses from the IV models showed that tissue ACE inhibitors as compared to non-tissue ACE inhibitors increased the hazard risk by approximately 30% to 60% (p < 0.05) for heart failure requiring hospitalization among the patients with heart failure and reduced hazard risk by approximately 30% to 40% (p <0.05) for AMI among patients without heart failure. In conclusion, though this study did not find significant difference between tissue and non-tissue ACE inhibitors for the comparative treatment effectiveness and safety in the study cohort, considerable comparative treatment effectiveness may exist in the subgroup of patients with and without heart failure in the elderly patients after AMI.

Dementia is a major public health problem worldwide. Emerging research indicates that clinical infections and PTSD could be important risk factors for dementia. However, evidence for infections and the risk of dementia primarily examines central nervous system (CNS) infections. Extant epidemiological evidence for systemic bacterial infections and the risk for dementia is limited while that for PTSD and the risk for dementia did not account for psychotropic medications commonly used in management of PTSD and could affect cognitive function. The purpose of this study was to 1) review the evidence for CNS infections as possible causes of Alzheimer’s disease (AD) dementia, and 2) using nationwide Veterans Health Administration databases, conduct original retrospective cohort analyses in nationally representative samples of U.S. veterans aged 56 years and older to determine the associations between systemic bacterial infections, PTSD, and psychotropic PTSD medication use with the risk for developing dementia. Review of the research pertaining to an infectious AD etiology hypothesis including the various mechanisms through which different clinical and subclinical infections could cause or promote the progression of AD, and the concordance between putative infectious agents and the epidemiology of AD showed evidence linking AD to an infectious cause to be largely inconclusive; however, the amount of evidence suggestive of an association is too substantial to ignore. Analysis of the associations between systemic bacterial infections and the risk for dementia showed a significant association between exposure to any systemic bacterial infection and an increased risk for dementia (hazard ratio [HR] = 1.20; 95% confidence interval [CI] = 1.16-1.24) after adjustment for demographic characteristics, and medical and psychiatric comorbidity. In addition, septicemia (HR=1.39; 95%CI=1.16-1.66), bacteremia (HR=1.22; 95%CI=1.0-1.49), osteomyelitis (HR=1.20; 95%CI=1.06-1.37), pneumonia (HR=1.10; 95%CI=1.02-1.19), UTI (HR=1.13; 95%CI=1.08-1.18), and cellulitis (HR=1.14; 95%CI=1.09-1.20) were independently associated with significantly increased risk of developing dementia after adjustment for potential confounders. Analysis of the associations between PTSD and psychotropic PTSD medication use with the risk for dementia showed a significant association between PTSD and the risk for dementia (HR=1.35; 95%CI=1.27-1.43) after adjustment for demographic characteristics, medical and psychiatric comorbidity, and health care utilization. Analysis of the impact of psychotropic PTSD medications including selective serotonin reuptake inhibitors (SSRI), serotonin-norepinephrine reuptake inhibitors (SNRI), benzodiazepines (BZA), novel antidepressants (NA) and atypical antipsychotics (AA) on the association between PTSD and the risk for dementia showed significant interactions between PTSD and use of SSRIs (p<.0001), NAs (p=.0016), and AAs (p<.0001). Multivariate analysis showed a significant association between PTSD and an increased risk for dementia among individuals not using any psychotropic PTSD medications at baseline (HR=1.70; 95%CI=1.58-1.82). PTSD patients using SSRIs (HR=2.10; 95%CI=1.82-2.41), NAs (2.19; 95%CI=1.94-2.48) or AAs (4.56; 95%CI=4.04-5.15) were significantly more likely to develop dementia compared to those without PTSD and not using any psychotropic PTSD medications. PTSD patients using SSRIs (HR=1.24; 95%CI=1.08-1.42), NAs (HR=1.29; 95% CI=1.14-1.46) or AAs (HR=2.69; 95%CI=2.38-3.04) were also significantly more likely to develop dementia compared to those with PTSD and not using any psychotropic PTSD medications. SNRI (HR=1.35; 95%CI=1.26-1.46) and BZA drug use (HR=1.40; 95%CI=1.35-1.45) at baseline was associated with an increased risk for dementia regardless of PTSD diagnosis. These findings indicate; 1) evidence for an infectious AD etiology hypothesis in inconclusive, 2) both severe (e.g. sepsis), and less severe (e.g. cellulitis) systemic bacterial infections are collectively and independently associated with an increased risk of dementia among older U.S. veterans hence prevention of systemic bacterial infections could positively influence the risk for dementia among older adults, and 3) PTSD and psychotropic medication use are associated with an increased risk for dementia among U.S. veterans. Further epidemiologic, clinical, and basic science research is required to elucidate the mechanisms and the associations between infections and the risk for dementia and to determine if the independent and effect modifying impacts of psychotropic PTSD medication use on the risk for dementia are related to differences in PTSD severity, other psychiatric comorbidity, or whether psychotropic PTSD medication use is an independent risk factor for dementia.

This dissertation focused on the topic of youth violence in two very different populations, young workers and youth offenders. Youth violence at the home, in school and in the community has been well documented in the literature but very little is known about the prevalence of and risk factors for victimization at the workplace in young workers. In chapter two, a retrospective cohort study was conducted using National Crime Victimization Survey data from 2008–2012. We calculated a rate of workplace violence victimization and compared those rates between occupations and demographic characteristics in young workers 16–24 years. Multilevel, weighted Poisson regression models were used to compare rates of workplace victimization across occupations and demographic characteristics. The rate of workplace violence victimization was 1.11 (95% CI: 0.95–1.27) incidents per 1,000 employed person-months. Young workers in retail sales occupations had a higher rate of workplace victimization than workers in health care occupations (RR = 0.52, 95% CI: 0.26–1.03) but a lower rate of workplace victimization than workers in protective service occupations (RR = 2.25, 95% CI: 1.34–3.77). Rates of workplace violence victimization differed significantly by age, income and workplace location. In contrast, the prevalence of and risk factors for juvenile offender are well-known. However, there still exist major gaps in determining the effectiveness of tertiary interventions, justice-based processes (i.e. formal appearance in court vs. informal agreement or meeting with court officer) and placement (e.g. detention centers, foster care, mental health institutions). In chapters three and four, the effectiveness of justice-based processes and placement on recidivism in young offenders 12–16.5 years were evaluated using data received from the Iowa Criminal and Juvenile Justice Planning Agency (CJJP) from 2010–2013. Multivariable logistic regression was used to impute risk level scores, to calculate propensity scores and to measure associations between demographic or complaint characteristics and main exposures. A sensitivity analysis was conducted by comparing the associations between process type and recidivism in a sample matched on propensity scores to the original unmatched sample. Cox Proportional Hazards models were used to compare time to recidivism by process type or placement in matched and unmatched samples. In chapter three, out of 2,901 youth offenders, 41% recidivated over an 18-month period. Eighteen percent were formally processed while 82% received an informal agreement. Youth who received an informal agreement had a lower risk of reoffending at any time compared to youth who were formally processed in both the unmatched (HR = 0.93, 95% CI: 0.76–1.13) and matched sample (HR= 0.86, 95% CI: 0.65–1.14). These estimates were not statistically significant. We observed an offense-specific association between processing and recidivism. Property (HR = 0.74; 95% CI: 0.57–0.96) offenders who received an informal agreement were significantly less likely to recidivate compared to property offenders who were formally processed. In chapter four, out of 1,469 youth offenders, 36% recidivated over an 18-month period. Nine percent received placement while 91% did not. Youth who received placement had up to an 87% higher risk of reoffending at any time compared to youth who did not receive placement in both unmatched (HR = 1.52, 95% CI: 1.11–2.08) and matched (HR= 1.87, 95% CI: 1.23–2.84) samples. We observed a charge-specific association between receiving placement and recidivism. Youth charged with a simple misdemeanor (HR = 2.74; 95% CI: 1.63–4.60) or other charges (HR = 6.60, 95% CI: 1.56–28.00) and received placement were significantly more likely to recidivate compared to those who did not receive placement. These findings contribute to the youth violence literature in the following ways. Chapter two identified the occupations and target populations in need of policies and evidence-based interventions aimed at improving the working conditions for young workers. Chapter three and four supports the continual evaluation of the juvenile justice system to determine the best practices that may reduce violence and recidivism in young offenders.