Skip to content
No Vape

A case-crossover study compares a person’s exposure across selected times

Find the event-related and reference periods without confusing self-matched observation with randomized crossover treatment.

No Vape · Updated

Stack of old newspapers on a table
Illustrative image · Mr Cup / Fabien Barral / Unsplash

Find who supplies the comparison

A familiar word such as crossover can hide different designs. Ask whether an intervention sequence was assigned, or whether the researchers selected periods around an observed event. In the latter case, the comparison can come from other times for the same person. Those reference periods are not another independently recruited group of people without the event.

Maclure’s original 1991 case-crossover paper and Marshall and Jackson’s original 1993 analysis describe this self-controlled observation design for transient exposures and acute events. We use the design distinction only. No disease example, effect estimate or claim that a particular vaping question suits this design is imported. This page supplies no instructions for testing exposure on yourself.

Maclure — American Journal of Epidemiology, original abstract via PubMed: The case-crossover design: a method for studying transient effects on the risk of acute events (1991) · Marshall and Jackson — Statistics in Medicine, original abstract via PubMed: Analysis of case-crossover designs (1993)

Draw times on one person’s line

Mark the defined event, the event-related exposure interval and each reported reference interval. Keep the actual dates, lengths and selection rules where available. If you have only an abstract that omits them, write windows unresolved. The drawing should reveal the comparison the authors made, rather than substitute an interval that seems intuitively more plausible.

In a fictional sketch, a study observes an event on one date and compares a measured exposure just beforehand with exposure in specified earlier reference periods for that person. Several time windows can still belong to one individual. This invented diagram does not establish an exposure effect, adequate window choice or a population event rate; it simply prevents a period count from becoming a participant count.

Keep stable identity separate from changing circumstances

The same-person arrangement holds identity constant, but time windows can differ in circumstances and measurement. Read the authors’ rationale and limitations rather than assuming that self-matching creates a randomized experiment. The exposure was not necessarily assigned by chance, and a person being their own referent does not explain why an event occurred.

Also distinguish the design from a descriptive case series. Cases can supply structured comparison periods in a case-crossover study, whereas merely listing their characteristics is a different task. The methods, not the presence of the word case, establish the comparison. Keep any reported association on its actual scale with the stated uncertainty instead of calculating a new risk from the selected windows.

Write a faithful design sentence

Your note can identify the event-defined participants, exposure measure, event-related window and reference-selection rule. Then name the main time-related limitation the source reports. If the material is too brief, seek the methods or preserve the missing detail. Do not declare a study conclusive or worthless merely because the same people appear on both sides.

This is introductory evidence reading, not a formal design appraisal or clinical recommendation. Seek expert interpretation when a health decision depends on the finding. Never alternate vaping, nicotine or treatments to reproduce a research comparison. The useful outcome is understanding what was compared and why a self-matched observation differs from both a personal experiment and a trial that assigns crossover sequences.

Sources and scope

Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.

  1. Maclure — American Journal of Epidemiology, original abstract via PubMed — The case-crossover design: a method for studying transient effects on the risk of acute events (1991)
  2. Marshall and Jackson — Statistics in Medicine, original abstract via PubMed — Analysis of case-crossover designs (1993)
  3. No Vape — No Vape — app features and store links

Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write