Group Study
D’Souza et al. Case-Control Study of Human Papillomavirus and Oropharyngeal Cancer. NEJM 2007
1. What is type of study design described in this article; prospective, retrospective, cross-sectional? What is a case-control study? Why did they use this type of study design? Describe advantages and disadvantages of a case-control study.
2. What is the rational to do this study? What is their hypothesis? Do they have one? Why don’t they state their hypothesis if they have one?
3. What is the study population? How do they select the “control”? Do you think there is any possible “bias” in case/control selection? Do they match the control, if so why and what would be the consequence of case-control matching? Do they use appropriate control if not who you think should be the “control”?
4. What is an odd ratio (OR) and a confidence interval (CI)? What does OR refer to? How would it be different than risk ratio (RR)? Why don’t they report RR?
5. How were the samples collected? What is the difference in sample collection between case and control? How would you recommend them to collect the sample? Were the techniques used in detecting HPV infection appropriate if not what technique you recommend?
6. From “Table 1” which factors may be confounding factors? How would you eliminate or reduce the effects of these factors? Do they miss any other factor(s)?
7. How would you interpret the results in “Table 2”? What is an adjusted odd ratio? Is there any particular sexual behavior associated with oropharyngeal cancer? Does “oral sex” have higher association to with oropharyngeal cancer than “vaginal sex”?
8. How would you interpret the results from “Table 3”? Why do serologic statuses of L1, E6, and E7 have higher correlation than oral infection?
9. Can you describe the relationship between tobacco/alcohol uses oropharyngeal cancer (Table 4)? What is synergy index? Do tobacco and alcohol uses increase the odd of having cancer in HPV+ group? If not, why?
10. Do the results show that HPV infection causes oropharyngeal cancer? Does this study demonstrate causality or association? Describe an ideal study you will conduct to prove if HPV infection causes oral cancer.