search this site.

220905P - CHOICE OF A STUDY DESIGN

Copyright 1995-2023 © by Prof Omar Hasan Kasule - Admin
Category:
Print Friendly and PDFPrint Friendly

Presented at a Webinar on Research Methodology in Health Sciences at Northern Area Armed Forces Hospital (NAAFH) on 5th September 2022. By Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Professor of Epidemiology and Bioethics


PRELIMINARIES OF STUDY DESIGN 1: IDENTIFYING THE POPULATION:

There are basically four ways of identifying populations for epidemiological study:

  • Geopolitical areas (village, town, district, etc.),
  • Lists of households,
  • Restricted or institutional such as special registers and schools, and
  • Health insurance databases.


PRELIMINARIES OF STUDY DESIGN 2: SELECTING THE POPULATION:

Four considerations are made in the process of population selection:

  • Representativeness,
  • Accessibility,
  • Accuracy of the population denominator,
  • The desired study size.


PRELIMINARIES OF STUDY DESIGN 3: SAMPLE SELECTION:

  • In reality, we study samples and not populations.
  • The samples represent the population.
  • The samples may be selected scientifically to be random or may be non-probability: convenience, quota, or purposive.


5 MAIN STUDY DESIGNS:

  • Observational studies
    • Cross-sectional study design
    • Case-control study design
    • Follow-up study design
  • Experimental studies
    • Randomized clinical trials
    • Randomized community trials


CHOICE OF STUDY DESIGN:

  • Each type of design is preferable in a given set of circumstances.
  • The cross-sectional design is the preliminary design and also provides health planning epidemiological data.
  • The case-control design is best for the etiological study of rare diseases. It is usually a preliminary study to be confirmed followed by a follow-up or experimental study.
  • The follow-up design is best for etiological studies because the time sequence is known. It is also preferred for common, severe, and easily observed diseases that only need a short follow-up.
  • The experimental design is used for the final proof of causality or efficacy of an intervention.