Epidemiology: What is Bias, Chance, and Confounders in Epidemiology?

#epidemiology #psmslectures #risk

The p-value is not enough!- in Epidemiology

In my previous post, we looked at the overview of study designs, which try to establish the relation between exposure and outcome, in other words, whether an association exists or not!


Does exposure always mean the outcome?

Sometimes there may be spurious association because the exposure and outcome is not measured using robust study designs like RCT. For example, there is the high rates of sunburns and also high rates of ice cream consumption during hot seasons. But eating ice cream does not lead to sunburn but this third-factor heat is responsible for both high ice-cream eating and sunburns.

Third factor: Tertium quid – Epidemiology

The error in establishing an association may be due to pure chance, bias, or confounders.

Bias is the systematic deviation from the truth due to faulty study design while confounders confuse us by intermingling with the actual cause, check the video below to understand it better.


Statistics and Effect Size in Epidemiology: Must-have tools!

Amidst the confusion created by confounders and other factors not obviously causing the disease, epidemiologist use statistics and derive conclusions using p-value or probability which measure chance! One step ahead of the p-value is measuring the effect size, how big or large the difference between different exposure and the outcome. For example smoking case cancer 20 times more than red meat (only 0.2 times). Hence, the p-value is not enough!

#confoundingfactors #epidemiology #researchmethods

study designs

Epidemiological Study Designs: An Introduction to Epidemiological Studies

Definition of Epidemiology

Epidemiology is the study of the distribution and determinants of health-related states or events in specified populations, and the application of the knowledge acquired to control disease in that population or geographical area.

what is epidemiology

Epidemiology definition is not just a definition but a philosophy

If suppose car seller uses this philosophy, he will look for the distribution and determinants, which increase the sales of car units, this is exactly what business analytics try to do to get more leads and increased sales of their products or services.

What do you mean by study in epidemiological study design?

In simple words, it is a scientific procedure involving answering questions using population data.

Now the way you collect data to increase or decrease the credibility of your answers. Now the way we collect data or information is termed as STUDY DESIGN more appropriately.

Ecological study and case-series

An ecological study is done on groups of people but lacks a comparison, so the latter in which specific people with the same exposure or disease are studied to describe something unusual.

as the human mind understands better when comparison or context has been put, so evolved the comparison studies.

epidemiological studies

Case-control and Cohort study

As we move from simple observations to comparative or analytic study, collected data becomes more valuable as in a case-control study we can calculate the odds ratio and tell the risk involved with the exposure in question. The same is true for cohort studies where we calculate the Risk ratio. A cohort study overcomes the limitation of a case-control study which is temporal association. (Whether the chicken first or egg!)

Randomised Control Trial (RCT): An interventional Epidemiological study

As analytical studies like cohort and case control are not free of bias and error people have found a way to balance it in both the group by randomized control trial (RCT) where study participants are randomly allocated to each group, and neither participants nor investigators know what intervention they are receiving known as double blinding.

Randomization is the heart of RCT, but we can only randomize people for intervention which is having benefits reported from initial observations, and never something potentially harmful to them.

Systematic review and Metanalysis: Summarize all the relevant studies done using RCT

Where systematic review synthesizes, analyzes, interprets, and summarises the findings, meta-analysis tries to combine all the statistical analysis done on similar studies.

Hierarchy of Epidemiological Studies

The hierarchy of study design exists because of our quest to find the most robust evidence which improves the quality of human life by reducing disease-related morbidity and mortality.

study design
study design vs analysis

Test your knowledge of Basic Epidemiology.

#epidemiology #studydesign #psmlectures