Opening your learning space
Starting the app …
Opening your learning space
Starting the app …
Log-Wald confidence intervals describe sampling uncertainty in a risk ratio or odds ratio from a 2×2 table.
Free to browse · flashcards or a multiple-choice quiz · import from Essential
Formula
a,b,c,d,n1,n0 are counts · RR, OR and interval bounds are dimensionless · z is a standard-normal quantile
| Symbol | Meaning | Unit |
|---|---|---|
| Events in the exposed or intervention group | people | |
| Non-events in the exposed or intervention group | people | |
| Events in the reference group | people | |
| Non-events in the reference group | people | |
| Exposed-group total a+b | people | |
| Reference-group total c+d | people | |
| z quantile for the chosen two-sided confidence level, e.g. 1.96 for 95% | dimensionless | |
| Point estimate of relative risk (a/n1)/(c/n0) | dimensionless | |
| Point estimate of the odds ratio ad/(bc) | dimensionless |
CI_{RR}=\exp\left[\ln(RR)\pm z\sqrt{\frac1a-\frac1{n_1}+\frac1c-\frac1{n_0}}\right],\;CI_{OR}=\exp\left[\ln(OR)\pm z\sqrt{\frac1a+\frac1b+\frac1c+\frac1d}\right]For a cohort 2×2 table, RR=(a/n1)/(c/n0) and OR=ad/(bc). Standard errors approximate ln(RR) and ln(OR); exponentiating both endpoints keeps intervals positive. These are asymptotic Wald approximations and may perform poorly with small or zero cells; exact or suitable score-based methods may be preferable. A confidence interval measures sampling uncertainty, not causality or clinical importance. In case-control designs, OR is generally directly estimable, whereas RR requires additional assumptions.
An independent 2×2 sample has adequate cell counts for the asymptotic Wald approximation.
Log-transformed ratio estimates receive symmetric limits on the log scale and are then exponentiated.
Rearrange for z quantile
For the symmetric log-Wald interval.
a=30,b=70,c=15,d=85. Calculate RR and its approximate 95% CI.
RR=2.00; SElog=0.283; exp[ln2±1.96·0.283]≈[1.15,3.48].
Leaving limits on the log scale or interpreting OR as RR.
Exponentiate both log limits and name the measure by study design; zero cells need other methods.
Formula cluster
a 95% interval above one is compatible with a positive association in the chosen group direction.
Table: a=30, b=70, c=15, d=85. RR=0.30/0.15=2.00; SE[ln(RR)]=√(1/30−1/100+1/15−1/100)=0.283, so the 95% CI is exp[ln(2)±1.96·0.283]≈[1.15,3.48]. OR=2.43 with SE=0.355 and CI≈[1.21,4.87]. Check: both positive intervals contain their point estimates.
Report uncertainty in epidemiological group comparisons from cohort and case-control studies
Here is how to work through a typical Confidence Intervals for Risk Ratios and Odds Ratios (Log-Wald CIs for RR and OR, then exponentiate) task step by step:
Task
a=30,b=70,c=15,d=85. Calculate RR and its approximate 95% CI.
Solution path
RR=2.00; SElog=0.283; exp[ln2±1.96·0.283]≈[1.15,3.48].
sampling uncertainty of an estimated RR or OR from a 2×2 table. Its output is lower and upper plausible ratio limits from a log-Wald approximation.. a 95% interval above one is compatible with a positive association in the chosen group direction. The quantity answers a specific biological or statistical question, not automatically the full cause of an observed pattern. Tie it to the population, sample or model level from which inputs came. A number without that reference can be misleading across studies. Keep description separate from explanation: the metric summarizes data but does not by itself establish the underlying mechanism.
Use a,b,c,d are cell counts, n1/n0 group totals and z the selected normal quantile. The units are counts are count values; RR, OR and bounds are dimensionless.. Check that numerator and denominator refer to the same time interval, target population and scale. Proportions entered as percentages usually need conversion to values between zero and one. Write units and definitions beside the observations. This helps catch swapped groups, mismatched age intervals or a wrong denominator before they distort the result. Also document measurement definitions and preprocessing, because studies can operationalize the same symbols differently.
The central assumption is: RR uses cohort risks or OR uses odds, with study design respected. the log approximation is reasonable with sufficiently large expected cell counts. If zero or sparse cells can distort Wald limits substantially., interpretation changes or the formula no longer fits. State sampling, measurement method and biological scale. A model is useful because it simplifies a question, but its limits belong in the report. Do not automatically transfer parameters to different species, populations, laboratories or sampling conditions. Report these assumptions with the result so readers can judge which populations and conditions the conclusion can cover.
In the example, 30/100 versus 15/100 gives RR=2.00 and approximate 95% limits 1.15 to 3.48. The check is: the interval remains positive and contains the estimate 2.00. Substitute calculated values back into the original relation and check whether ranges and units are biologically plausible. Rounded data can create a small discrepancy, but should not change direction or order of magnitude. Also test a simple limiting case, such as zero, complete detection or equal groups. This separates arithmetic errors from model limitations. Round intermediate values only at the end, and check the order of magnitude against the biological question.
A nearby quantity is the Wilson interval for one proportion. The substantive distinction is that Wilson bounds one proportion p; these log formulas address group ratios RR or OR. Both can appear in the same biological analysis, but they use different inputs and answer different questions. Choose by the requested output and verify the data definition before comparing values. A related name does not imply identical units or interchangeable interpretation. This matters when a descriptive statistic is reported alongside a model parameter. Before comparing values, align definitions, populations, observation periods and analysis rules; otherwise the difference may be methodological.
Quanta exam set
Curated exam set for "Confidence Intervals for Risk Ratios and Odds Ratios":
Question (front)
Which formula describes Confidence Intervals for Risk Ratios and Odds Ratios?
Answer in your set
Question (front)
How do you rearrange Confidence Intervals for Risk Ratios and Odds Ratios for Rearrange for z quantile?
Answer in your set
Question (front)
Which common mistake happens with Confidence Intervals for Risk Ratios and Odds Ratios?
Answer in your set
+ 8 more cards: units, variables, derivation, example, exam task
The formula set contains 11 cards. After importing, review them and start practicing; your ratings determine later review intervals.
This list names subject references and further reading. A page or section number appears when it is part of the reference; these are not verbatim quotations.
Create a curated FSRS exam set for Log-Wald CIs for RR and OR, then exponentiate: formula recall, variables, derivation, rearrangement, worked example, common mistakes and exam context.
Curated formula set · LaTeX · FSRS spaced repetition · import from Essential