Self-controlled case series
Does event incidence vary within people around a transient exposure?
When to use this project
Use a self-controlled case series for a transient exposure and an acute event when reliable exposure, event and observation dates are available among people who experience the event. Compare defined risk and baseline periods within each person.
How it differs
A cohort or case-control comparison uses differences between people. SCCS controls factors constant within a person during observation, but time-varying factors still need attention. It differs from case-crossover sampling and from a descriptive clinical case series.
Decisions before you start
- Assess whether an event changes later exposure or ends observation.
- Prespecify risk, pre-exposure, washout and baseline periods.
- Specify recurrence assumptions, age and calendar-time effects; use an appropriate extension when standard assumptions fail.
How the flows work together
Each flow below links to its diagram, steps and tables. Flow order organizes the project; it does not imply that all work is strictly sequential. Data assessment, analysis planning and protocol refinement may inform one another.
1. Self-controlled study overview
Overview · Question through within-person incidence comparison
Use this overview to state the purpose and link to detailed decisions. Keep a versioned protocol and decision register, and add evidence references for the rationale.
2. Case and event selection
Selection · Identify cases and qualifying events
SCCS includes people who experienced an event. Check that event occurrence does not invalidate the exposure or observation process assumed by the chosen model. Keep event counts separate from case-person counts.
3. Observation and exposure windows
Data · Partition baseline and risk periods
Partition each person’s observation into nonoverlapping intervals with recorded durations. Define risk windows before inspecting associations and address pre-exposure periods, repeated exposures and age or calendar effects.
4. Within-person analysis plan
Statistics · Specify incidence contrasts and adjustments
Conditioning within people removes factors constant during observation, not time-varying confounding. Use a model with interval duration and appropriate age or seasonal structure; pooled event proportions do not reproduce SCCS.
5. Assumptions and sensitivity
Statistics · Assess event-dependent exposure and observation
Events may delay exposure or end observation. Assess those mechanisms and recurrence assumptions; use a suitable extension or reconsider the design. Testing alternative windows does not repair every assumption violation.
Equations and statistical interpretation
These explanations describe notation and assumptions. Flow does not calculate these values.
Conditional rate model notation
Nᵢⱼ is the event count and Tᵢⱼ the duration in person i’s interval j. X indicates a risk period; Z describes time-varying factors. Conditioning on each person’s total removes the person-specific αᵢ; exp(β) is the relative incidence for the specified contrast.
Do not estimate the SCCS effect by simply pooling risk-period and baseline counts across people. Standard model assumptions about event-dependent exposure, observation and recurrence require assessment; extensions must match the violation.
Published example
Kwong et al. (2018), Acute Myocardial Infarction after Laboratory-Confirmed Influenza Infection
Inspect the post-infection risk window and control intervals used within individuals. The study concerns infarction after infection; it is not an estimate of influenza vaccine effectiveness.
This publication illustrates a study design, not an endorsement of every method or a reproduction of its results. Follow its protocol, supplement and update history when studying the example.
Hypothetical scenario
A safety team wants to assess whether an acute cardiac event occurs more often shortly after a documented infection than during other observed periods in the same people.
- Self-controlled study overview: Define the infection, acute event and within-person contrast.
- Case and event selection: Identify cases and confirm observation periods independent of event occurrence where required.
- Observation and exposure windows: Partition each person’s follow-up into prespecified exposure and baseline intervals with age and calendar time.
- Within-person analysis plan: Fit a suitable conditional event-count model using interval durations.
- Assumptions and sensitivity: Assess event-dependent observation or exposure, recurrent-event dependence and alternative risk windows.
Deliver a populated protocol, decision log, data specification, and planned outputs. Enter numerical findings only after analysis and verification. Keep decisions made before seeing results distinguishable from later amendments.
Sources
- Petersen, Douglas and Whitaker (2016): self-controlled case series methods
- RECORD checklist for studies using routinely collected health data
- HMA–EMA Catalogue of RWD studies: registration and protocol guidance
- Whitaker et al. (2006): Tutorial in biostatistics—the self-controlled case series method
Source check: 2026-09-17. Record the version of each reporting or appraisal tool used in your actual protocol.
Compare all Observational and RWE studies project types · Open editor in a new tab