150 likes | 176 Views
Explore the significance of prospective and retrospective studies in determining causation, common design flaws, and the importance of critical analysis in research. Gain insights from the smoking and lung cancer case study.
E N D
Experimental Design Group 2: Nabilah, Jing Kai, Soon Guan
Prospective Study vs Retrospective Study • Flaws in Studies • Cause and Effect • Case Study: Smoking and Lung Cancer • Conclusion Outline Experimental Design
Reading Salsburg, David (2001). The Lady Tasting Tea: How Statistics Revolutionized Science in the Twentieth Century, W.H. Freeman (Chapter 18, Does smoking cause lung cancer?)
Prospective Study • Subjects are selected and monitored over a period of time for development of disease Problems: • Extrapolation issues • Time consuming
Retrospective Study • Subjects with disease are identified • Investigations for prior conditions associated with disease Problems: • Rare diseases
Flaws in Studies • Every study has its flaws • Non-random samples → Reinforcement of results through consistency over many studies
Cause and Effect: A implies B = Not A implies not B
Postulates for Causation • Whenever the agent can be cultured, the disease is there • Whenever the disease is not there, the agent can’t be cultured • When the agent is removed, the disease goes away
Does Smoking Cause Lung Cancer? Case Study
Overview • Prospective & retrospective studies done • Widely reported that smoking causes lung cancer
Postulates not met • Flaws in experimental design • Publication bias (Fisher, 1958) • Confounders Eg. Genetics
Conclusion • Correlation does not imply causation • Not easy to determine cause and effect • Studies are still important as they can still give an idea and provide clues on the possible underlying causes of a disease