Analysis and missingness plan
Specify estimates and attrition handling
Repeated measurements from one person are dependent. Address measurement timing, attrition and clustering, and distinguish descriptive trajectories from causal exposure comparisons.
Flow diagram
Steps
Specify estimates and objectives
Define target population, outcome summary and any comparison.
Justify information requirements
Choose precision or size rationale appropriate to the objectives.
Specify analysis methods
Address repeated observations, clustering and relevant dependence.
Plan missing-data handling
Distinguish intermittent missingness, dropout and informative follow-up.
Assess attrition and sensitivity
Compare plausible missingness and censoring assumptions.
Document assumptions and limits
Separate planned specifications from observed findings.
Analysis and missingness plan
End flow
Tables and supporting information
Analysis specifications
Add rows for each outcome and comparison.
Missingness and sensitivity
Address loss to follow-up explicitly.
Survival function
T is time from the specified origin to the event, and S(t) is the probability of remaining event-free beyond t. Estimation requires a censoring assumption appropriate to the method. For a particular cause in the presence of competing events, use a cumulative incidence estimand rather than interpreting one minus cause-specific Kaplan–Meier survival as its risk.