<p> Small-study effects, which are factors resulting in dependencies between treatment effect size and precision, are an important source of bias in meta-analyses of randomized controlled trials. However, established nonparametric tests for detection of small-study effects that are based on rank correlation lack statistical power, while established parametric tests that are based on linear regression are not robust in the presence of between-study heterogeneity. </p><p> A novel method for detection of small-study effects is proposed that is designed to overcome these limitations. The method uses repeated one-sample Wald-Wolfowitz runs tests to evaluate the null hypothesis of serial independence among trial treatment effect size estimates that are ranked by precision. This dissertation describes lower-tailed, upper-tailed, and two-tailed versions of the proposed method for detection of small-study effects and compares the proposed method to established tests using simulation. The novel method is implemented in Stata using various procedures for control of type 1 error, including the Bonferroni and Sidak corrections, Hochberg&rsquo;s step-up procedure, and the Benjamini-Hochberg procedure for control of the false discovery rate. The type 1 error rate and power of the novel method are then compared to those of existing tests, including the nonparametric rank correlation test of Begg and Mazumdar and the commonly-used regression-based tests of Egger, Harbord, and Peters. Factors known to affect the performance of established tests, including effect size, number of trials in each meta-analysis, degree of between-study heterogeneity, and degree and type of publication bias (a specific cause of small-study effects) are simulated to reflect characteristics of meta-analyses in the biomedical literature. </p><p> The simulation demonstrated that all of the procedures evaluated for control of type 1 error in the novel method maintained an error rate below the nominal rate under all scenarios, suggesting that any of these procedures may be used to implement the novel method. In contrast, error rates for the established tests of Begg and Mazumdar, Egger, Harbord, and Peters were at or above the nominal rate under most scenarios. The lower-tailed, upper-tailed, and two-tailed novel tests showed little power in excess of the type 1 error rate under all conditions. In contrast, established tests demonstrated variable power depending on the conditions. Specifically, the power of established tests increased with an increase in effect size, an increase in the number of trials in each meta-analysis, an increase in the severity of publication bias, and publication bias that operated by effect size rather than by <i> p</i>-value. In contrast, the power of established tests decreased with an increase in heterogeneity. Overall, Egger&rsquo;s test demonstrated the highest power. Despite the low power of the novel method, selected circumstances under which it may be useful are described.</p>

<p> <b>Introduction:</b> During 2017, approximately 750,000 strokes occurred, of those approximately 140,000 people died from a stroke and &lt;10% were also diagnosed with a hospital acquired condition. The economic burden associated with stroke care exceeds $34 billion; and hospitalizations with a hospital acquired condition increase costs. In the United States, differences in health outcomes have been documented; however, these studies show that sociodemographic differences continue to exist. The objectives of this study are to assess differences in clinical outcomes among the study population, and to assess if there are differences hospital charges. </p><p> <b>Methods:</b> To investigate differences in outcomes and hospital charges among the study population, the 2001-2011 Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS) database was used. Hospitalizations that occurred with a primary diagnosis of an ischemic stroke and a subsequent diagnosis of a hospital acquired condition were queried using ICD-9 codes. Additionally, the Elixhauser comorbidity index was used to identify comorbid conditions among individuals hospitalized. To assess the differences in outcomes and hospital charges chi-square, logistic regression, and hierarchical multilevel modeling procedures were used. All statistical analyses were performed using SAS 9.4. </p><p> <b>Results:</b> The overall death rate has decreased among the study population; however, racial/ethnic differences exist in patient outcomes. When assessing hospital charges, hospitalizations that occurred in the Southern region of the United States were higher than hospitalizations that occurred in the Northwest region of the United States. </p><p> <b>Conclusion:</b> Further analysis needs to be conducted to assess sociodemographic differences in clinical outcomes among the study population. There is a need to continue to identify sociodemographic groups with risks of mortality to better guide the funding opportunities to target these resources to populations that experience the worse health outcomes. Future studies should also assess the role of mental health, cultural competency, and care coordination to improve patient outcomes.</p><p>

<p> <b>Background.</b> Prostate cancer is the most common non-cutaneous cancer and the second leading cause of death among men in the United States. It highlights one of the highest racial disparities in health outcomes across cancers, with Non-Hispanic Black (NHB) men being at a 1.6 times higher risk of being diagnosed and 2.5 times higher risk of dying from the disease compared to Non-Hispanic White (NHW). Stage at diagnosis is the major metric used for prostate cancer prognosis, and assesses the extent of the disease. Prostate cancer development and progression is a multifactorial phenomenon, influenced by factors ranging from biological interactions to nativity. Therefore, we aimed at describing the characteristics of prostate cancer cases in New York State, and conduct an exploratory analysis to determine among the factors commonly associated with prostate cancer development and progression, which sociodemographic determinants contributed to racial disparities in stage at diagnosis of prostate cancer in New York State. To do this, we accessed the New York State Cancer Registry (NYSCR). </p><p> <b>Methods.</b> Prostate cancer cases recorded in the NYSCR from 2004 to 2014 constituted our sample. The NYSCR report prostate cancer stage in both the American Joint Committee on Cancer (AJCC) Tumor Node Metastasis (TNM) classification system and the National Cancer Institute Surveillance Epidemiology and End Results Program (SEER) summary staging 2000. For this study, the SEER Program summary staging 2000 was used. We created mutually exclusive categories for each covariate (grade at diagnosis, race, age at diagnosis, insurance status, marital status and nativity), while accounting for the sociodemographic landscape of New York State. Descriptive statistics, univariate and multivariable logistic regression models were computed, adjusting for grade at diagnosis, race, age at diagnosis, insurance status, marital status, and nativity. The variables found to have more than 10% missing data were removed and the data reanalyzed, to see the influence of data quality on our results. The Hosmer-Lemeshow test was used to assess the goodness of fit of each model. </p><p> <b>Results.</b> The sample consisted of 164,765 cases with a mean age of 66.72 &plusmn; 9.79 years. The greatest proportion of cases was diagnosed as local or regional stage prostate cancer. Statistically significant associations with stage at diagnosis in both the adjusted and unadjusted models were found with grades 3 and 4, NHB, Hispanics, patients aged 65-74, 75-84 and more than 85 years old, Medicaid and Medicare insured patients, patients with no insurance, patients with unspecified/other insurance, patients not married, patients with unknown marital status and patients of whom nativity status was unknown. The association between insurance status and distant stage at diagnosis was statistically significant for patients with military insurance in the unadjusted model, but not after adjustment. Birthplace information was missing for 42.53% of our sample. After birthplace was removed from the model and the data reanalyzed, the association for military insurance became statistically significant, while the association for unknown marital status became non-statistically significant. </p><p> <b>Conclusion.</b> Within this population-based sample of New York State prostate cancer cases, stage at diagnosis was found to be associated with grade, race, age at diagnosis, insurance status, and marital status. The multiplicity of factors associated with distant stage at diagnosis confirms the multifactorial nature of disparities in prostate cancer outcomes. The high percentage of missing data precluded the accurate assessment of the role of nativity. However, the factors identified here to act on racial disparities in stage at diagnosis in New York State provide a solid foundation for future analysis.</p><p>

<p> Objective: To investigate the impact of the density of foreign born Asian population on geographical incidence rate of liver cancer in New York City during 2009 &ndash; 2013 in order to find specific geographic areas in NYC where liver cancer intervention should be targeted. Method: We chose to employ cross-sectional and ecologic study design. We collected count data for liver cancer cases and sociodemographic characteristics from the 2010 U.S. census tracts (n = 2120) and health indicators from the United Hospital Funded neighborhoods (n = 34) in New York City during 2009 &ndash; 2013. We performed multilevel analysis in order to investigate the association between the density of foreign born Asian population and geographical incidence rate of liver cancer, controlling for sociodemographic characteristics at the census tract level and health indicators at the UHF neighborhood level. Result: We found that for each one-percentage increase in the foreign born East Asian population in a census tract region, there is a significant increase in the expected incidence rate of liver cancer by 1.0%, controlling for other variables. Conclusion: There was significant impact of the density of foreign born East Asian population on geographical incidence rate of liver cancer in NYC. We expected that the UHF neighborhoods with relatively high density of foreign born East Asian population and high liver cancer incidence rate should be targeted for the public health intervention of liver cancer.</p><p>