Independent Poisson rate comparison worksheet
StatMedAnalysis | Version 2026-11-05

Instructions: save a working copy, fill blank fields, retain assumptions and
record an independent arithmetic check. Use fictional or aggregate values;
do not enter patient identifiers. This is a planning aid, not clinical advice.

1. Define comparable numerators and denominators
Event definition and whether repeat events count: _________________
Group 1 population, event count and total exposure: _______________
Group 0 population, event count and total exposure: _______________
Shared exposure unit and follow-up rules: _________________________
Censoring, entry and time-at-risk exclusions: _____________________

2. Audit assumptions before selecting an interval
Are groups independent? Can events cluster within people or sites?
Is fixed known exposure reasonable? Any evidence of overdispersion?
Are counts sparse or zero? If so, plan an appropriate exact or score
method; this worksheet's log-Wald interval is not a sparse-data remedy.
Confounding variables and adjusted analysis plan: _________________

3. Recompute
The log-Wald calculation requires both event counts and exposures > 0.
Rate1 = events1/exposure1: __________ Rate0: ______________________
Ratio = Rate1/Rate0: _____________________________________________
SE(log ratio) = sqrt(1/events1+1/events0): _________________________
Approximate 95% interval = exp(log(ratio) +/- 1.959964 SE): _________
Hypothetical example: 30/1000 versus 20/1000 gives ratio 1.5,
log-SE sqrt(1/30+1/20)=0.288675; interval approximately 0.851865–2.641265.

4. Report and review
Rate multiplier (e.g. per 1000 exposure units): ____________________
Unadjusted or adjusted, interval method and limitations: __________
Clinical/contextual interpretation: ______________________________
Do not describe a rate ratio as a risk ratio or infer causality from it.
Prepared by/date/version: __________ Independent check: __________

Guide and primary-source links (verified 2026-11-05):
https://www.statmedanalysis.com/blog/poisson-rate-ratio-audit
https://www.statsmodels.org/stable/generated/statsmodels.stats.rates.confint_poisson_2indep.html
https://www.statmedanalysis.com/tools/poisson-rate-ratio