Bias in Epidemiological Research
Epidemiology is the study of the factors associated with different types of diseases for example, how often does the disease occur, how is the disease transmitted, ways in which the disease can be prevented. There are two main types of epidemiology: descriptive epidemiology and analytical epidemiology. Descriptive epidemiology is concerned with the frequency and distribution of risk factors in a population and it makes it possible for one to assess the way the disease has spread.
Analytical epidemiology aims to study the causes and risks involved with the disease and the preventive measures.
This is according to Pinchinat, S, & Ponton Sanchez (2006) Observation epidemiological studies In this case the epidemiologist does not carry out any actual experiment or in other words does not take part in any active role in the research. The epidemiologist simply makes observations on what is happening based on an already existing situation without administering any treatments for example if one wanted to compare the exposure of PCBs exposure to occurrence of cancer over a 20 year period, he would not be required to give any form of treatment only to observe the available case available according to epidemiology home page.
Experimental epidemiological studies In contrast to observational studies here the epidemiologist will be required to do the actual experiments or play an active role by administering the treatment to subjects and then observing the effects of the treatment. For example an epidemiologist can perform a clinical trial of a new drug on willing subjects and then observe the changes in subjects based on epidemiology home page. There are different types of biases associated with epidemiological research.
According to Eric’s Notebook, bias arises when an estimated value deviates from the original or true value. The case studies given in the assignment are going to provide a basis for the discussion of various types of biases and the effects of the biases on the measures of association. Based on CES-Research-bias & confounding, different types of biases will be discussed. Case 1 This case might show both selection and measurement biases.
Selection because the epidemiologist might select the children non-randomly in which case the results could be overestimated or underestimated in that if he over picked children who are not exposed to the chemical then his or her result will be underestimated and vice versa. The results will show a bias in measurement if the epidemiologist measures the outcome inaccurately in which case the results will either be over or underestimated. Case 2 This case could show measurement and analytical biases. Analytical bias is evidenced when patients give false information for example about having less sexual partners.
This would lead to over or underestimation of the results. Measurement bias would occur if the epidemiologist collected the wrong data from the studies. This also would lead to over or underestimation of the results. Case 3 This case will show two kinds of biases; measurement bias because the epidemiologist may collect inaccurate data and analytical bias because of not following up the outcome. All this would lead to either under or overestimation of data. Case 4 Measurement and analytical biases could occur in this case.
Incase of analytical bias the results could remain unchanged or underestimated and in case of measurement bias it could either be under or overestimated. Word count: 551 References Aschengran, A, & Seage, G. R, (2008). Essentials of epidemiology in public health, (2nd Ed. ). Sudbury, MA: Jones and Bartlett. CEM-Research- Bias & confounding is a site that provides essential information on research bias on epidemiology. file:///c:/biasconfound. html Checkoway, H, Pearce, N, Kriebel, D. (2004) Research methods in occupational epidemiology.
Epidemiology Home Page is a website that gives an introduction to epidemiology File:///c:/epi1. html. Ibrahim, A. M. (2001) Eric Notebook: Selection Bias. Department of veterans’ affairs, epidemiologic research, and information center at Durham, NC. http://eric. unc. edu/notebooks/issue8/eric_notebook_8. pdf Pinchinat, S, & Ponton-Sanchez, A, (2006): Analytic and descriptive epidemiology. Biostatem SARL. File:///c:/epidemiology. htm Study Types in Epidemiology http://www. nwcphp. org/training/courses-exercise/courses/study-types-in-epi