“Simpson’s paradox, or the Yule–Simpson effect, is a phenomenon in probability and statistics, in which a trend appears in several different groups of data but disappears or reverses when these groups are combined. It is sometimes given the descriptive title reversal paradox or amalgamation paradox.”
This seems counterintuitive, but the 5 minute video below explains the concept well.
Wikipedia: Simpson’s paradox
Minute Physics: Simpson’s Paradox
A set of training materials for professionals working in intervention epidemiology, public health microbiology and infection control and hospital hygiene.
JAMA: Physician Payments from Industry Are Associated with Greater Medicare Part D Prescribing Costs
Oxford Reference: A Dictionary of Epidemiology (6 ed.)
American Journal of Epidemiology: A Dictionary of Epidemiology, Sixth Edition Edited by Miquel Porta
NCBI: Association of Facebook Use With Compromised Well-Being: A Longitudinal Study.
Stat: NIH rejected a study of alcohol advertising while pursuing industry funding for other research
This is some brief guidance from my advisor on how to write the introduction section of an epidemiology scientific paper. When addressing previous papers in the introduction, do so only briefly. Generally, save the thorough literature review for the discussion.
What is the public health or clinical importance of the topic? What is the primary problem that will be addressed? How many people will be affected? What level of impact does this problem have? Statistics from the World Health Organization are often cited here.
What is currently known about the problem?
For example, what has been published on health related quality of life (HRQOL) in type 2 diabetes mellitus (T2DM) patients?
Briefly describe a variety of primary literature papers on the topic. State the lacking knowledge that will be addressed by the rest of the paper.
There is much known about HRQOL in T2DM in populations of White Americans, but there have been no studies to date describing HRQOL in Pacific Islanders diagnosed with T2DM.
Address challenges unique to this study.
Are there variations in HRQOL perceptions among different cultures?
Clearly and concisely state the primary aim of this study.
For example, in the current analysis we will study the impact of T2DM on HRQOL in a population of Pacific Islanders living in Oahu, Hawaii.
Say something specific about the population being studied.
The Pacific Islander Cohort of Hawaiians is a longitudinal, population-based cohort that has been ongoing since 1999, with followup every 4 years.
Explain why this study is novel. Tell what you are going to show.
Hemoglobin A1c (HbA1c) is a validated clinical measure of T2DM severity (citation here), and the SF-36 is a validated health questionnaire measuring HRQOL (citation here). To the extent of our knowledge, this is the first study to study a potential quantitative association between HbA1c and the SF-36 in a population-based cohort of Pacific Islanders.
Mendeley is a convenient, free research resource that allows you to manage primary literature references. Mendeley’s Citation Plugin allows easy citations in Microsoft Word while drafting scientific papers from your library.
