Project Design
Based on the goals of your project, you will need to select an appropriate study design.
Often when starting a study/project that involves statistics, you will need to decide on the type of study you are conducting. You should understand the three main categories of study design. They are:
Descriptive
A descriptive study design is one in which your primary goal is to assess a sample at one specific point in time without trying to make inferences or causal statements. In general, there are three primary reasons to conduct descriptive studies:
- To identify areas for further research.
- To help in planning resource allocation (needs assessment).
- To provide informal information about a condition or disease.
Descriptive studies are helpful in revealing patterns and connections that might otherwise go unnoticed. We will discuss two types of descriptive studies:
- case reports
- and cross-sectional.
Descriptive Case Report Study
A descriptive case report study is a detailed profile of a subject or set of subjects and their clinical or laboratory experience. The purpose of these studies is to identify potential areas of research and generate interest and hypothesis formulation.
Case reports are a first step for providing clues to identifying a new disease or adverse health effect from an exposure or clinical experience. A classic example is the case report study of 5 homosexual males who developed a rare pneumonia. This case report study led to the eventual discovery of HIV. Case reports are most likely to be useful when the disease or condition is uncommon and when the disease or condition is caused almost exclusively by a single kind of exposure.
Descriptive Cross-Sectional Study
A descriptive cross-sectional study is a study in which the disease or condition and potentially related factors are measured at a specific point in time for a defined population. Cross-sectional studies can be thought of as a “snapshot” of the frequency and characteristics of a condition in a population at a particular point in time.
This type of data can be used to assess the prevalence of conditions in a population. However, since exposure and disease status are measured at the same point in time, causal statements should not be made from this type of study. It is usually not possible to distinguish whether the exposure preceded or followed the condition.
Retrospective
In a retrospective study, the outcome of interest has already occurred at the time the study is initiated. A retrospective study design allows the investigator to formulate ideas about possible associations and investigate potential relationships, although causal statements usually should not be made.
An investigator conducting a retrospective study typically utilizes administrative databases, medical records, or interviews with patients who are already known to have a disease or condition. In general, the reasons to conduct a retrospective study are to:
- Study a rare outcome for which a prospective study is not feasible.
- Quickly estimate the effect of an exposure on an outcome.
- Obtain preliminary measures of association.
We will discuss two types of retrospective studies:
- case-control
- and cohort.
Retrospective Case-Control Study
In a retrospective case-control study the investigator can quickly estimate the effect of an exposure on outcome status. Cases and controls are established based on the presence of the condition, and exposure is assessed by looking back over time. It is especially important in a case-control study that the cases be as similar to the controls on all factors except the outcome of interest.
Retrospective Cohort Study
A retrospective cohort study allows the investigator to describe a population over time or obtain preliminary measures of association to develop future studies and interventions. The exposure and outcome information in a cohort study are identified retrospectively by using administrative datasets, reviewing patient charts, conducting interviews, etc.
Prospective
Prospective study design is the best design for establishing relationships between your outcome of interest and exposure variables. The primary feature of prospective designs is the outcome has not occurred at the time the study is initiated, and information is collected over time to assess relationships with the outcome.
We will discuss:
- prospective cohort studies
- and randomized controlled trials.
Prospective Cohort Study
In a prospective cohort study, subjects are identified based on exposure status prior to the development of the disease or condition of interest, and the outcome is measured over time. In this type of study, the investigator can establish a temporal relationship between an exposure and an outcome, or follow the natural development of a condition over time.
Randomized Controlled Trials (RCTs)
Randomized controlled trials (RCTs) are considered the gold-standard of study design. They can provide evidence for causal relationships and support changes in clinical practice. In an RCT, subjects are randomly assigned to receive the intervention or control treatment, and outcomes are evaluated after the intervention period. The control group is the group which receives the standard of care (or a placebo).
It is especially important that the investigator and subjects are blinded to which treatment the subjects receive. This is called a double-blind study. Conducting a double-blind study helps to reduce any potential bias in the results of the study.