Observational Study Designs: Synopsis for Selecting an Appropriate Study Design.

Assad A Rezigalla

doi:10.7759/cureus.6692

Abstract

The selection of a study design is the most critical step in the research methodology. Crucial factors should be considered during the selection of the study design, which is the formulated research question, as well as the method of participant selection. Different study designs can be applied to the same research question(s). Research designs are classified as qualitative, quantitative, and mixed design. Observational design occupies the middle and lower parts of the hierarchy of evidence-based pyramid. The observational design is subdivided into descriptive, including cross-sectional, case report or case series, and correlational, and analytic which includes cross-section, case-control, and cohort studies. Each research design has its uses and points of strength and limitations. The aim of this article to provide a simplified approach for the selection of descriptive study design.

Highlights

BackgroundA research design is defined as the “set up to decide on, among other issues, how to collect further data, analyze and interpret them, and to provide an answer to the question” [1]
The primary objective of a research design is to guarantee that the collected evidence allows the answering of the initial question(s) as clearly as possible [2]
Each of them deals with the specific type of research or research questions and has points of strength and weakness

Summary

Introduction

A research design is defined as the “set up to decide on, among other issues, how to collect further data, analyze and interpret them, and to provide an answer to the question” [1]. The assessed associations are guided by sound hypotheses and seen as hypothesis-generating [17] This design can be descriptive (when dealing with prevalence or survey) or analytic (when comparing groups) [17,18]. It starts with the outcome of interest (referred to as cases) and looks back in time for exposures that likely caused the outcome of interest [13, 20] This design compares two groups of participants - those with the outcome of interest and the matched control [12]. Can study the association of risk factors and outcomes in outbreak investigations It can generate much information from relatively few participants with unusual cases. Lack of bias in the retrospective cohort because the collected data was not initially for research It can measure potential causes and relative risk. Two crucial points that should be noted during the process selection include different study designs that may be applicable for the same research question(s) and researches may have grey areas in which they have different views about the type of study design [4]

Disclosures

24. Setia MS