Description of Public Health Research Study
The published public health report selected for this paper is a study by Shegog, Lindley, Thompson-Robinson, Simmons, and Richter (2010) aimed at identifying HIV and STI risk factors among African-American students attending predominantly white higher institutions of education. The study emerges from statistics that show that most African-American students study in predominantly white universities. Although they face major social challenges that lead to HIV/STI, such as drug abuse, like other races, few public health interventions address their unique social issues, such as racism and alienation, and their impact on HIV/STI incidence in this demographic (HealthyPeople.gov., 2012; Galvan, Davis, Banks, & Bing, 2008).
Research Question and Methods Used
The research question employed by Shegog et al. (2010) in their secondary research study stated that “What behavioral risk factors lead to a positive HIV/STI diagnosis among African-American students in predominantly white universities?” As mentioned above, the authors of this report used a secondary research study to answer the research question by employing a backward stepwise logistic regression method which evaluated factors that could lead to a positive HIV/STI diagnosis among the selected population. The authors developed their findings by analyzing secondary data from ACHA – NCHA Spring 2007 data (Shegog et al., 2010). This data pool included responses from students regarding their sexual behaviors, alcohol and drug use, condom use, and the prevalence or incidence of HIV and sexually transmitted diseases throughout their college lives. The authors also examined 26 items from the selected data to identify the ones that could lead to a positive HIV/AIDS diagnosis. Shegog et al. (2010) also analyzed variables that had many responses by using a dichotomous response model to predict HIV/STI risk. Lastly, they also conducted the step-wise logistic regression analysis using SAS version 9.1.3 (Shegog et al., 2010).
Appropriateness of Study Design and Methods
The use of the backward stepwise regression method was appropriate for the study because the authors already had a selected group of variables that they had to eliminate to remain with those that could reliably predict a positive HIV/STI diagnosis. This method allowed them to have a pool of variables that would have a statistically significant effect on their analysis. The use of the secondary research analysis in answering the research question was also a reliable metric for investigating the validity of past primary research studies that often explored the relationship between condom use, or drug use, and HIV or STI incidences among African-American college students (Shapiro, Mostashari, Hripcsak, Soulakis, & Kuperman, 2011). In other words, the findings of the study helped to understand the effects of primary research on the selected topic.
Additional Insight and Validation of Idea
Although we have established the rationale for using the backward stepwise logistic regression model in the assigned article, it is important to verify the findings of the researchers by investigating the model’s accuracy in the study. To do so, I propose that future researchers should not rely on the F-statistic and multiple R of the regression model to investigate the same health topic. Similarly, they should not predominantly focus on the model’s significance and ignore their limitations (Shi & Johnson, 2014). Instead, researchers should evaluate the model based on a set of data that did not primarily create the body of research used in the initial study. This idea is valid because the accuracy of some regression models has been often questionable because of errors. By adopting this proposal, we could identify some predictors of HIV/STI selected by Shegog et al. (2010) that could be flawed.
Galvan, F., Davis, E., Banks, D., & Bing, E. (2008). HIV stigma and social support among African Americans. AIDS Patient Care and STDs, 22(5), 423–436. Web.
HealthyPeople.gov. (2012). DATA2020, the Healthy People 2020 interactive data tool. Web.
Shapiro, J. S., Mostashari, F., Hripcsak, G., Soulakis, N., & Kuperman, G. (2011).
Using health information exchange to improve public health. American Journal of Public Health, 101(4), 616–623. Web.
Shegog, M., Lindley, L., Thompson-Robinson, M., Simmons, D., & Richter, D.
(2010). HIV/STI Risk Factors Among African-American Students Attending Predominantly White Universities. Journal of Health Disparities Research and Practice, 4(1), 86–98.
Shi, L., & Johnson, J. A. (Eds.). (2014). Novick& Morrow’s public health administration: Principles for population-based management (3rd ed.). Burlington, MA: Jones & Bartlett.