Written by MicroDok

Experimentation is an active attempt to change a variable in a group of people. For example, the trial of a particular drug which may be new can be done or carried out using a defined population in order to determine the potency or efficacy of the drug. This type of experimentation in testing a trial drug is critical before the drug is released into the open market for public use. For example, a population of 20 people is used and they are divided into 2 groups as: Population 10A and Population 10B. The 1st group (Population 10A) is given the active drug while the 2nd group (Population 10B) is given a placebo (i.e. drug totally different from the active drug). A placebo is a tablet which appears to be a drug, but contains no medicinal substance in it. Placebos are used on control groups in tests of new drugs. The 1st group serves as the test population and they have the disease condition that the drug is supposed to act upon or cure while the 2nd group is the control group and they are free from the disease condition (even though they received the inactive part of the drug). In this way, the potency/therapeutic efficacy of the trial drug is determined.

Ethical Consideration is very important in carrying out experimental epidemiological studies. And the need for ethical consideration is aimed at:

  • Maintaining patient’s confidentiality in the course of the experimentation/research.
  • Avoiding a bridge of the patient’s fundamental human rights or infringing into it.
  • Not subjecting the patient to conditions that is detrimental to his or her health.
  • Protecting the researcher from any contamination during research.
  • Not obtaining relevant data from the patients under duress

Ethical clearance must be obtained from the ministry of health or hospital where the work is to be carried out in order to avoid bridge of patient’s confidentiality or other health policies. It is also very important that the techniques the epidemiologists or researcher is experimenting on and the laboratory procedure to be used for the study must be an already accepted procedure or criteria that must be generally acceptable by the international community as something that is workable.


  1. Randomized Control trial: This is an investigation in which subjects are randomly selected and allocated into different groups. Participants in such a study are usually divided into the test group and the control group. It is usually done to assess an experimental therapeutic agent in a given population. The results are assessed by comparing the health experience/status of the test group and that of the control group in the study. The result should be subjected to statistical analysis in order to check if there is any significant difference in the results obtained or not. If there is any significant difference, then it implies that the results obtained before was not by chance but if there is no significant statistical difference, then it imply that what happened before was by chance.

There are two categories under the randomized control trials:

  1. Single Blind trials: In single blind trials, the person administering the drug knows which patient gets the active/real drug, and those that gets the placebo. But the patient does not know whether he or she is getting the active drug or the placebo.
  2. Double Blind trial: In double blind trials, neither the investigator nor the patient knows who gets the placebo or the active drug. Placebo is not the real drug. It is a substance used in epidemiological studies; and such substances or chemicals have little or no therapeutical value. Placebos are generally inert in nature, and confer no therapeutic value to the study population taking them. Placebos only appear to be the real medical treatment; but the reverse is the case, because they are not the real drug and varies from the actual drug being tested in the study.
  3. Confounding variables: These are variables that can cause or introduce bias into the study. These variables can also introduce errors in the study or reduce its accuracy. They are very important both to descriptive and experimental studies because in every experiment it may be difficult to keep every variables at constant. There are some variables that can be controlled and some that cannot be controlled.

Controlling Confounding Factors in epidemiological studies

            The ways by which confounding factors can be controlled in an epidemiological study include:

  1. Randomization
  2. Restriction
  3. Matching
  4. Stratification
  5. Multi-variant analysis

Randomization: Randomization is simply the process of conducting or sampling at random. Random sampling is usually conducted without any specific consideration of the individuals that make up the population during sampling. Sampling at random helps the researcher to include the various representative members that make up the population under study. Here, the people that make up the population are screened without a prior knowledge of where they are coming from; what they do for a living; where they work or what they eat.

Restriction: Restriction as a means of controlling confounding variables in an epidemiological research is mainly aimed at reducing bias in the work. Because a bias may be introduced, the work is restricted to individuals who have similar characteristics within the population under study. Here, the individual’s socio-economic class (occupation, education, residence) is considered. An example of such a measure is shown in Table 1.

Table 1. Prevalence of sexually transmitted infection in Enugu (A)

Occupation No examined No (%) infected
Civil servants

Okada riders




Here, your intervention study is restricted to a particular research. Instead of saying prevalence of sexually transmitted infection (STIs) in all of the above occupation, it can be restricted as thus: Prevalence of STIs amongst Okada riders in Enugu metropolis.

Stratification involves obtaining certain socio-demographic parameters from the study population because of difficulty in defining and restricting the study to a particular group of the population. Some of these socio-demographic parameters includes: occupation, sex, religion, tribe, marital status, residential location, education level.

Table 2. Prevalence of sexually transmitted infection in Enugu (B)

Occupation                 No examined              No (%) infected

 Okada riders               100                                          50(50%)

Students                      50                                            5 (10%)

Farmers                       60                                            25(41%)

Traders                        40                                            10(25%)

Civil servants              50                                            5(10%)

Total                           300                                          95(31.7%)

For interventional purposes when using stratification model, you don’t work with the final result (31.7%) but rather with the individual results as shown in Table 2. 

Matching: In matching technique, the study population must match exactly with the control population to be used. For example, if the study population is students, then the control group in the study must also come from student population. It will be a misdemeanor to use a different control group and different test group for the study. Also, the occupation of your test population must also match the occupation of your control population if the socio-economic factor you have decided to study is the occupational aspect of the population – as it relates to the disease under investigation. Matching helps the researcher to eliminate error in the results.

Multi-variant analysis: This involves subjecting your results/findings to statistical analysis. This includes using Chi-Square (x2), regression and correlation analysis, analysis of variance (ANOVA) and T-test. In order to know if the result was by chance or not; and to test the significance of the results obtained, it is important to include multi-variant analysis in the study in order to eliminate bias. Multi-variant analysis helps in knowing the interventional measures to be used in the study. Statistical analysis is also important in controlling confounding factors and it goes ahead to prevent any possibility of doubt in the work or results obtained from the study.

In disease prevention and control; two research approaches are usually considered:

  1. Social research
  2. Scientific research – is based on experimentation carried out in the laboratory on animals or other non-human or human models to ascertain the cause of a disease outbreak in a bid to contain its further spread.

Social Research: Social research entails or involves knowing the social aspect of the community where the disease outbreak occurred. Here, the knowledge of the community about the disease outbreak is ascertained. Their attitude and practice is also ascertained. Different techniques are used in carrying out the social research and they include:

  • Identification of the socio-demographic parameters of the community. These socio-demographic parameters include the age, sex, religion, occupation, education, and the geographical and climatic factors of the disease outbreak. Social research helps in obtaining information that cannot be obtained in the laboratory. A holistic manner/approach is used in going about this, before a preventive and control strategies are put in place to contain the disease spread. In social research, a preliminary investigation is carried out in order to get a base line, so that your work will not be based on what people say. Social research has two basic areas; and they are
  1. Qualitative survey
  2. Quantitative survey

Qualitative survey involves doing interviews using an interview guide. This interview guide will contain a list of questions that the test population will be asked. In qualitative survey, there is a person-person interaction in which the investigator talks with the test population on individual basis; and the answers obtained from such a one-to-one interview are recorded by the investigator. A group discussion in which the investigator becomes the moderator can also be put up in the qualitative survey.

Quantitative survey involves the use of a structured questionnaire or pro-forma. Here, specific questions including the name, age, sex, occupation and educational status of the interviewee (i.e. the person to be interviewed) are typed in a paper and given to the person to fill and submit later. After which, the questionnaires are obtained back from the people that received them; and they are finally analyzed, interpreted and implemented in the prevention and control measures of the disease outbreak.


Salyers A.A and Whitt D.D (2001). Microbiology: diversity, disease, and the environment. Fitzgerald Science Press Inc. Maryland, USA.

Schneider M.J (2011). Introduction to Public Health. Third edition. Jones and Bartlett Publishers, Sudbury, Massachusetts, USA.

Singleton P and Sainsbury D (1995). Dictionary of microbiology and molecular biology, 3d ed. New York: John Wiley and Sons.

Slonczewski J.L, Foster J.W and Gillen K.M (2011). Microbiology: An Evolving Science. Second edition. W.W. Norton and Company, Inc, New York, USA.

Songer T (2005). Study designs in epidemiologic research. Supercourse, (http://www.pitt.edu/~super1/lecture/lec19101/index.htm) (Accesed May 2103).

Stedman’s medical dictionary, 27th edition. Philadelphia: Lippincott, Williams and Wilkins.

Summers W.C (2000). History of microbiology. In Encyclopedia of microbiology, vol. 2, J. Lederberg, editor, 677–97. San Diego: Academic Press.

Taylor LH, Latham SM, Woolhouse ME (2001). Risk factors for disease emergence. Philos Trans R Soc Lond B Biol Sci, 356:983–989.

Willey J.M, Sherwood L.M and Woolverton C.J (2008). Harley and Klein’s Microbiology. 7th ed. McGraw-Hill Higher Education, USA.

Aschengrau A and Seage G.R (2013). Essentials of Epidemiology in Public Health. Third edition. Jones and Bartleh Learning,

Aschengrau, A., & G. R. Seage III. (2009). Essentials of Epidemiology in Public Health.  Boston:  Jones and Bartlett Publishers.

Biosafety in Microbiological and Biomedical Laboratories. 5th edition. U.S Department of Health and Human Services. Public Health Service. Center for Disease Control and Prevention. National Institute of Health. HHS Publication No. (CDC) 21-1112.2009.

Bonita R., Beaglehole R., Kjellström T (2006). Basic epidemiology.  2nd edition. World Health Organization. Pp. 1-226.

About the author


Leave a Comment