Revision notes · Lecture 5 · Free

Descriptive study designs

In one line: descriptive studies answer who, where and when, never why. They describe one group, with no comparison group.

What this lecture's questions test

Out of the 30 questions on this lecture:

IdeaQuestions
Cross-sectional studies13
Case reports and case series8
Ecological studies5
Descriptive or analytical?4

1. Descriptive or analytical?

  • Descriptive: person (who), place (where) and time (when). It cannot answer why, and it cannot identify risk factors.
  • Comparing two groups makes a study analytical.
  • Measuring incidence needs a disease-free group followed over time. That is a cohort study, not a descriptive one.

2. The four descriptive designs

DesignStudiesBest for
Case reportOne patientA rare or new presentation, an unexpected drug effect
Case seriesA group of similar patientsUncommon clinical features. No comparison group
Cross-sectionalIndividuals, at one point in timePrevalence, burden, generating hypotheses
EcologicalGroups: districts, states, countriesRelating disease rates to exposure rates, generating hypotheses

Remember it: one patient → report. Several → series. A snapshot of a population → cross-sectional. Whole populations compared → ecological.

  • A case report cannot give incidence: there is no population to divide by.

3. Cross-sectional studies

  • Exposure and outcome are measured at the same time: a snapshot.
  • The unit of observation is the individual. It counts all existing cases, old and new, not only new ones.
  • Uses: burden (prevalence), trends, generating hypotheses. It can measure several outcomes at once. A census is a cross-sectional survey.
  • Main limitation: it cannot show which came first (temporality), so it cannot establish cause.
  • The analytical cross-sectional study compares groups within the snapshot, and so can identify factors associated with an outcome.

Remember it: association yes, sequence no.

4. Ecological studies

  • The data are group-level: exposure and outcome for each district or country, never for each person.
  • Strength: cheap and quick, using data that already exist. It relates rates and generates hypotheses.
  • Weakness: it cannot link one person's exposure to that person's outcome. Assuming it can is the ecological fallacy.
  • Example: tobacco sales and oral cancer rates, state by state.

Watch out: one true/false question asks whether a descriptive cross-sectional study has a comparison group. The key and standard teaching disagree; the app explains it on that question.

Last look before the exam

  1. Descriptive: who, where, when, not why.
  2. One case → report. A group → series. Neither has a comparison group.
  3. Burden or prevalence → cross-sectional.
  4. Cross-sectional cannot show temporality.
  5. Group-level data → ecological.

Practise lecture 5’s 30 questions