7312MED-Epidemiology

Task

Read the paper by Bond, Egan, Kearns, Clark and Tannahill (2012) (required reading) and answer the following questions.

1.What type of epidemiological study design was applied in this study? Identify two key features of this type of study design in the given paper? 

2.The authors have established two research questions (p.300) in relation to their research aims. According to these two questions, state two research hypotheses. 

3.What is the study factor and how was this factor measured? 

4.What is/are outcome factor(s)? How was the factor/ were they measured?

5.Do the main findings presented in Table 2 suggest any significant association between the study factor and the outcome factor? What do the results suggest? 

6.Could selection bias have possibly been introduced into the study? What effects could this bias have had on the observed association between the study factor and the outcome factor(s)? Provide reasoning to support your answers. 

7.Is information bias likely to have influenced the results? Discuss your view briefly. 

8.How have the authors addressed the issue of confounding? What are the potential confounders identified by the authors? With regard to Table 2, have the confounding factors influenced the association between the study factor and the outcome factor? Why or why not?

9.According to the results shown on Table 3, have the authors answered their second research question (p.300)? Why or why not?  (Hints: examine OR, 95%CI changes, combining with P values to discuss the authors’ summary on p. 302, the end of RESULTS section)

10.As illustrated by the authors, the smoking rates in both HI and non-HI groups were much higher than the national norm of 26% for men and 23% for women. Do you agree with the authors that high smoking rates observed in the study population were due to chance or factors not related to receiving HI? Why? Explain your reasons briefly.

11.Which of the Bradford Hill’s criteria for evaluating causality can be established through this study? Give one reason for each criterion of your choice.

12.To whom may the results be generalised?