From a methodological perspective, clinical research can have an experimental or an observational design, which translates into experimental or observational trials or studies, respectively. In an experimental design, the investigator establishes the treatment, selects the patients, indicates what variables will be measured, etc. In an observational study, investigators can only observe the events without interfering or modifying the conditions of routine clinical practice.
Observational studies can be classified as:
- Case reports
- Case series
- Cohort studies
- Case-control studies
Classification of Observational Studies:
Descriptive Studies: They answer questions such as “who,” “where” and “when,” describing events as an approach to reality. Descriptive studies include case reports and case series.
- Case Reports: They are detailed studies of one or more occurrences, usually less than 10, which comprise about 30% of original published articles in journals. Case reports are used to describe rare findings and, in general, they serve to formulate a hypothesis that will later have to be tested in other studies. Since it is a reduced group of patients and due to the selection method, case report conclusions are usually biased.
- Case Series: They are similar to case reports but they involve a larger number of patients (usually more than 10) with a certain adverse reaction or disease and exposure. In general, investigators try to establish a relationship between the disease and exposure using statistical evidence but without a control group for comparison.
Analytical Studies: They answer the question "how." These controlled studies try to know reality by exploring reasons and causes and checking hypotheses. The most common are cohort studies and case-control studies.
- Cohort Studies: They are used to study incidence, causes and prognosis. Because they measure events in chronological order they can be used to distinguish between cause and effect. Cross-sectional studies are used to determine prevalence. They are relatively quick and easy but do not allow to distinguish between cause and effect. In this type of study, groups of exposed and non-exposed patients (control group) are selected and allocated to a certain treatment, which is the object of the study. When they are used to assess the safety of a treatment, a follow-up over time is conducted to analyze whether patients present or not with a certain adverse reaction, disease or, in general, an event that may have a negative impact on health. Cohort studies are usually prospective and, less frequently, retrospective. Follow-up is usually long to allow time for the disease or the adverse reaction under study to occur. Furthermore, if this reaction is rare, the sample size of the cohorts must be very high. This type of design is the most appropriate to study relatively common events that may occur after short exposure to treatment or when treatment is rare.
- Case-Control Studies: Two groups of patients are selected: a case group, which includes patients with a certain adverse reaction (disease, harmful effect, etc.), and a control group, which includes randomly selected patients of the same population as the case group but who do not show that adverse reaction. Then, both groups are studied for a certain time (window of exposure) prior to the beginning of the adverse reaction or exposure to the risk factor under study (drug or treatment). Its design is retrospective because, at the time of study start, the events have already occurred. This design is the most appropriate when the adverse reaction under study is rare or highly unusual.
Observational studies are mainly useful to increase knowledge about drug safety but, on the other hand, they present several disadvantages, such as lack of efficiency, since they are very expensive and the patient sample is usually relatively reduced and highly homogeneous; therefore, in general their results have little external validity.