Research Study Design

Objective- To devise a study method that will clearly answer the study question with the least amount of time, energy, cost, and patient risk.

Timing of Observation Will patients be examined once, or followed over time, allowing for multiple outcome assessments?

Timing of Observations Only one observation of the risk and outcome variables. Cross-sectional Study

Timing of Observations Multiple observations of risk and outcome over time. Cohort (Longitudinal) Study

Timing of Study Question Development Will the data be collected from a time period that precedes or follows the development of the study question?

Timing of Question Development Question proposed prior to data collection and outcome. Prospective Study

Timing of Question Development Question proposed prior to data collection but after the outcome occurrence. Retrospective Study

Cross-Sectional Survey Provides a snapshot view at one instant in time of the relationship between risk and outcome.

Incidence A rate at which people develop a given disease or condition over a specific period of time. (Ex: Each year, 1000 cases of AIDS occur.)

Prevalence The proportion of a people who have a given disease at one point in time. It is not a rate, although it is often called a prevalence rate. (Ex: Today, 1,000 people have AIDS)

Point Prevalence The same as prevalence, since prevalence is recorded at one point at time.

Cross-Sectional Study Optimal design when causality not previously suggested. Risk/outcome relationship suggested, but not proven by this design.

Case-control Study The retrospective comparison of patients with different outcomes (diseased, not diseased), with a retrospective look at risk variables.

Case-control Study Matching: Selection process that insures that cases and control subjects are similar with respect to confounding factors. Prevents adverse influence on true risk/outcome relationship.

Case-control Study Analyze risk variables in both outcome groups. Determine if risk is unequally distributed in the cases and controls.

Matching The process of selecting study and control subjects so that they are similar with respect to confounding factors.

Confounding Variable A factor that is related to both the risk variable and outcome that mediates a change in the apparent association between the exposure and disease (ie: suggesting an unreal association or masking a real association.

Confounding Variable A variable more likely to be present in one group of subjects than another that is related to the outcome of interest and thus potentially confuses, or “confounds,” the results.

Nested Case-control Study A case-control study that includes patients who are originally part of a cohort study.

Risk Variable Manipulation Will the researchers observe or intervene?

Observational Study The study of risk variables and outcome without alteration of any risk variables by the researcher.

Interventional Study The study of outcome that involves the manipulation of one or more risk variables by the researchers.

Observational Study A study that does not involve an intervention or manipulation. It is called case-control, cross- sectional, or cohort, depending on the design of the study.

Observational Study A study which describes risk variables and outcome. Descriptive

Observational Study The study of risk variables and outcome which attempts not only to describe the risk variables and outcome, but also attempts to determine their relationship. Analytic

Descriptive Statistics Statistics such as mean, standard deviation, proportion, and rate used to describe attributes of a data set.

Randomization A procedure for assigning treatment to each patient in such a way that it is equally likely (within the constraints of the experimental design).

Randomization Patients are allocated to the treatment and control groups by chance.

Intervention Plan Will the interventions be randomly assigned?

Intervention Delivery Patients are randomly assigned to different experimental groups (ie: placebo vs. drug). Randomized Trial

Intervention Delivery Patients are not randomly assigned to different experimental groups (ie: placebo vs. drug), but instead are provided different interventions under the control of the researchers. Non-randomized Trial

Research Awareness Do the data collectors know to which experimental groups patients have been assigned?

Blinding To deprive of perception or judgment.

Blind Study An experimental study in which subjects do not know the treatment patients are receiving; investigators may also be blind to the treatment patients are receiving.

Single-blinded Trial Patients do not know to which group they have been assigned.

Double-blinded Trial Patients and investigators do not know which group each patient has been assigned.

Double-blinded Trial A clinical trial in which neither the subjects nor the investigator(s) know which treatment subjects have received.

Open Trial A trial in which both the patient and investigator know that they are receiving the study drug (intervention).

Prophylactic Trials Trials designed to find ways to prevent the occurrence or spread of disease.

Therapeutic Trials Trials designed to find better ways to treat established diseases.