Epidemiological Study Designs

• Dr. Surangi Jayakody (MBBS, MSc, MD (Public Health), Assistant Professor at [email protected] presents the lesson.
• The lesson is about study designs and how to identify them.

Learning Outcomes

• Explore the different epidemiological study designs.
• Appreciate the strengths and weaknesses of each design.
• Identify the appropriate design for a given research problem.

Levels of Evidence

• Hierarchy of Scientific Evidence (Strongest to Weakest): Meta-analyses & systematic reviews > Randomized controlled trials > Cohort studies > Case-control studies > Cross-sectional studies > Animal trials & in vitro studies > Case reports, opinion papers, and letters.
• Not Scientific Evidence: Youtube videos, personal anecdotes, gut feelings, parental instincts, Natural News, Info Wars, Natural Health Warriors, Collective Evolution, Green Med Info, Mercola.com, Whale.to, etc.

Study Designs

• All study designs have similar components: a defined population (P) from which groups of subjects are studied, and outcomes (O) that are measured.
• Analytic studies also include: Intervention (I) or Exposure (E) that is applied to different groups of subjects.
• An illustrated chart outlines the relationships between Study Designs (Q1), Descriptive, Analytical (Q2), Experimental, Observational (Q3), Qualitative studies, Case reports, Case series, Correlational studies, Cross-sectional studies, Randomised Control, Quasi-experimental, Cross Sectional Analytical, Case-Control, and Cohort.

Spotting the Study Design

• Q1: What is the aim of the study?
  • Describe a population (Descriptive) uses PO questions.
  • Quantify the relationship between factors (Analytical) uses PICO questions.
• Q2: IF Analytical, did the researcher allocate the intervention?
  • Yes? the study design is RCT (Randomized Controlled Trial).
  • No? the study design is Observational.
• Q3: When were the outcomes determined?
  • A period after the exposure or intervention? The study is a cohort study.
  • At the same time as the exposure or intervention? The study is a cross-sectional analytical study.
  • Before the exposure was determined? The study is a case-control study.

Descriptive Studies

• Describe the patterns of disease occurrence in relation to variables such as person, place and time.
• Often the first step or initial enquiry into a new topic, event, disease or condition.
• Emphasize features of a new condition or describe the health status of communities or populations.
• Descriptive studies answer What? Why? Where? When? and Who? relevant questions.
• Descriptive studies are useful to help generate a hypothesis, but can't confirm causative/risk factors.
• Individual level includes: Case report, Case series, and Cross sectional study.
• Population level studies: Correlational / Ecological study.

Case Report and Case Series

• Describes a single patient or group of patients with similar diagnosis.
• A clinician identifies an unusual case/ unusual feature of a disease.

Cross Sectional Studies (Prevalence Studies)

• Surveys of the situation existing in a given population at a given time.
• Information obtained relates to a single specified time.
• Can also collect historical data.

Correlation / Ecological Studies

• Population level/ study unit is a group.
• Characteristics of entire populations are used to describe a disease in relation to a factor of interest (income, sex, utilisation of health services, diet).
• Is the first step in investigating a possible exposure disease relationship (eg. Diet and cancer).
• Uses available data - quick and inexpensive.
• The degree of association does not reflect an individual-level relationship.
• Limitations: Data represent only averages complex relationships cannot be determined, confounding, and changes associated with exposure can be masked.

Analytical Studies

• Are carried out to find the answer to "Why."
• Test hypothesis and have a comparison group.

Case-Control Design

• Starts with disease (Cases); the control group: are those without the disease.
• Both groups should be at the same risk of being exposed to the factor of interest.
• Advantages: Quicker and less expensive than cohort, useful to investigate rare disease, enables study of several different aetiological factors, has no attrition problem and minimal ethical issues.
• Disadvantages: Recall bias: has to depend on memory or past records to determine exposure, is difficult to select an appropriate control group, and cannot measure the incidence.

Cohort?

• A cohort is any group of individuals who are linked in some way or who have experienced the same significant life event within a given period.
• The investigator selects a group of exposed individuals and a group of non-exposed individuals and follows up both groups to compare the incidence of disease (or death rate).
• Determine a temporal relationship
• There is no randomization of exposure (no manipulation of exposure).
• Often require long follow-up period.
• Types of cohort studies: prospective cohort, retrospective (historical).

Experimental Studies

• Clinical trials are done on patients with disease.
• Prevention trials – on healthy people/people at risk of certain conditions.
  • Field trials – eg on immunization assigned to individuals.
  • Community trials- assigned to group of people.
• Clinical trials- Basic steps: Formulate a hypothesis, define eligibility criteria for experimental group, obtain consent, randomization (study group and control group), assign intervention to the study group, follow up for a predetermined period, and ascertainment of outcome.
• Randomly assign subjects to a treatment, i.e., an “exposure and compare incidence
• RCT.

Group Work

• Identify the study design.
• Define the study population.
• Define the outcome.
• Define the intervention or exposure.