Community trial
- Study design
A prevention programme is tested in whole communities (towns, neighbourhoods) and compared with others that do not receive it.
What it is
In a community trial the unit of intervention and analysis is a whole community: a municipality, a neighbourhood, a rural district. It is used for programmes that act on the whole population (vector control, media campaigns, changes to the environment) and cannot be given to some people while withheld from their neighbours. Its great limitation is that there are usually very few communities, so chance does not balance them and results are imprecise.
How it is built
- Communities similar in size, disease burden and services are chosen.
- They are allocated, ideally at random and in similar pairs, to programme or control.
- The outcome is measured before and after in all of them, with the same surveillance.
- The change in programme communities is compared with the change in control communities.
Worked example
- 8 towns with dengue are randomised: 4 get a community programme against breeding sites and 4 carry on as usual.
- Mean incidence before: 30 per 10 000 in programme towns and 28 per 10 000 in control towns.
- One year later: 18 per 10 000 versus 25 per 10 000. Change: −12 in the programme and −3 in control.
- Difference in changes: −12 − (−3) = −9 cases per 10 000; rate ratio after = 18 / 25 = 0.72.
- But with only 4 towns per arm, an outbreak in one of them would shift the mean a lot.
- Reading: the programme seems to lower incidence, with large uncertainty because there are few communities.
Community trial and neighbouring designs
| Aspect | Community trial | Cluster trial | Ecological study |
|---|---|---|---|
| Unit | Whole community | Group (school, clinic) | Population or region |
| Investigator assigns the intervention | Yes | Yes, at random | No: observes only |
| Usual number of units | Few (4 to 20) | Dozens | Variable |
| Individual data | Sometimes | Yes | No (aggregate) |
| Main limitation | Few units and contamination | Design effect | Ecological fallacy |
Markers
- Community unit
- Rate ratio
- Contamination
When to use it
- Vector control programmes, water fluoridation, anti-tobacco or healthy-eating campaigns.
- Interventions that rely on community leaders, local radio or changes to the environment.
- Useful for evaluating policies before extending them to a whole province.
Common pitfalls
- Using only one community per arm: any difference specific to those places is confused with the effect.
- Analysing people as independent, ignoring that the unit is the community.
- Forgetting that programme messages can reach control communities through media or migration.
Test yourself
Think of the answer, then tap the question to see it.
What is the unit of allocation in a community trial?
The whole community (municipality, neighbourhood or district), not the person.
Why do few towns limit a community trial even if thousands live in them?
Because communities are what is randomised: with few units chance does not balance their differences and the estimate is imprecise.
How does it differ from an ecological study?
In a community trial the investigator assigns the intervention; in an ecological study they only observe existing aggregate data.
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: Chapter 23: Including variants on randomized trials (Cochrane, 2024); Consort 2010 statement: extension to cluster randomised trials (EQUATOR Network, 2012); Principles of Epidemiology | Lesson 1 - Section 7 (Centers for Disease Control and Prevention, 2012).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.