For the purposes of study and scientific research, researchers have employed many different types of study designs. These are as follows: A. Case series This is the simplest and may be stated as most basic type of study design. Here, the study is simply just a description of a case that is observed. This describes the patient’s presentation, characteristics and the ‘story’ that unfolds as the patient’s case is handled. There is no group used for comparison and no control group. It is simply a case that is described in total for further reference or to gain ideas for more studies in the future.
The danger for bias in this study however is that wrong associations may be made due to the lack o a control group. There may be several factors pr plausible cases within the case and these may be misleading in terms of the conclusions made in the case observation. As such, the use of this case series is limited. In order to avoid bias, it must be merely used as an observation study or a reference to be used in the monitoring of new patients with seemingly unusual cases. B. Ecological studies The next type of study design would be the ecological study. Here, unlike the case study, there is comparison between the subjects at hand.
What makes it different however from other comparative studies is that here in this study, there is a comparison between groups of individuals rather than just one individual to another. The danger of bias in the case however happens when there is an association that is found on both a group level and an individual level. The problem here is that when an association is made, the researcher can be misled to make a conclusion on a group level, when in fact it is attributable to the people on an individual level. Should this happen, it is called ecological bias.
A good example of this would be a study made between religion and suicide. Say for example it was seen that there are a higher amount of cases of suicide in areas inhabited by a larger percentage of Religion X. To state that people of the said religion are more likely to commit suicide would be bias as we do not know if those who committed suicide are in fact of that religion. It’s possible that the association may not be associated to the group of people from that religion, but instead to small minorities of individuals in the population who so happen to live in the areas inhabited by people of Religion X.
In order to avoid this bias, careful decision making must be employed as to when ecological studies are to be employed. If used well, they have their benefits as they are cheaper and quicker, and may be conducted with routinely published information, such as birth and death rates. C. Cross sectional study The cross sectional study is a study that focusing on gathering information from a study population at a certain point in time. This would be like taking an instant snapshot at one point in time, one being a survey of the characteristics of the population under study at that particular point in time.
Here, the participants are taken from a well-defined study population. The characteristics of the participants are withdrawn only focusing on the aspects of that population at that one point. An example of this would be the exit polls made during voting day. Here, the population at hand would be those who went to vote at that one voting area. They are only contacted once, and the information gathered is based merely on what is found out about them at that one particular point that the information is gathered. These studies are cheap as people are only contacted once.
The weakness of this type however is that it can only detect the prevalence of a disease, but it cannot give information as to when they were first contracted and how long the disease has been present. Also, the behavior of people may change once they get a disease. This cannot be accounted for by the study as it only detects the presence of the disease at that one point in time, but not its onset beforehand or its effect afterwards. D. Case-control studies In this study, the study population is chosen primarily based on the disease that they possess. The importance of this process is that the control selection must be done with care.
For example, a study is to be made with respect to the association of smokers and risk for colon cancer. The study will focus on looking for people who had colon cancer and then check which percentage of these people were smokers. As you see, the population was chosen based on their possession of the disease at hand, namely colon cancer. Care now lies on selecting good candidates. For example, it would imprudent to choose people who have lung cancer, emphysema or heart issues, as these people are more likely to be smokers and it will cause misleading results.
One will not know where the colon cancer may be associated to the smoking or not, as there are other smoking-related factors in the mix of information. Another type of bias is recall bias, which can occur if the controls and the cases recall the previous events in different ways. Say for example, if the people interviewed regarding their disease do not remember the details of their disease so well, then it’s very possible that the information garnered from them is not as accurate as what occurred in the real life scenario. The strength of this study is for rare diseases.
This ensures that we are able to find an adequate number of people for the disease study. For example, instead of getting a group of smokers and seeing how many will develop the disease, if they even do develop it, this study starts with locating those who already have the disease, and from that point making the association with smoking. With this study, one can do research on rare diseases and ensure that there is enough an amount of study population. E. Cohort studies For this type of study, there is a healthy group of people that are identified.
These people are then followed up over time and it is referenced as to whether or not they develop the disease. This type of study is able to give information regarding the very onset of the disease and its progression thereafter. It is a prospective study. The important factor to prevent bias is that prior to its onset, neither the researchers nor the test subjects should have any idea who will be getting the disease, or the odds of such. This will prevent recall bias and other biases related to the onset and time factors in the study. The down side of this study is that it can be very expensive.
For example, if one were to study cancer development in people exposed to a chemical, one would have to follow up on the test group up until they develop cancer, a time frame that could take years. F. Randomized controlled trials This study is what is termed as the gold standard of all studies in epidemiology. This is an experiment wherein one group is assigned to receive one treatment and the other is not. Here, the people under study are not provided any information as to whether or not they are the ones in the control group or the treatment group.
If possible, the researchers as well are not given this information. The goal of this trial is to simulate a true-to-life situation, so that the results are gained in a fashion that is closest to what would happen in a real-life scenario. These trials are expensive and are only applicable in some cases, as humans cannot be used as test subjects when there is possible harm involved. Nevertheless, it is these tests that can bring us the most realistic information and are most beneficial in the field of research epidemiology.