Dunson · Human reproduction (Oxford, England) 1999 · Secondary statistical analysis of prospective cohort studies · n=?

Day-specific probabilities of clinical pregnancy based on two studies with imperfect measures of ovulation.

Cited 237 times in the scientific literature.

Level 3 - non-randomized controlled study

Secondary statistical modeling and re-analysis of observational cohort data

PubMed 10402400 · doi:10.1093/humrep/14.7.1835 · record verified 2026-08-26

What was done

Researchers applied a statistical model to correct for measurement error in identifying the day of ovulation and re-estimate day-specific fecundability and the duration of the fertile window. The model was applied to two historical datasets: a study of Catholic couples using natural family planning in London (1950s–1960s) who tracked ovulation via basal body temperature shifts, and a study of couples attempting pregnancy in North Carolina (early 1980s) who monitored ovulation via urinary hormone assays.

What was found

The abstract reports no exact numerical probabilities. After controlling for measurement error, both datasets demonstrated an identical 6-day fertile interval. In both cohorts, the estimated probability of clinical pregnancy peaked on the day immediately prior to ovulation and dropped to near zero following ovulation.

Why it matters

Shows that differing estimates of the fertile window across historical studies were likely artifacts of measurement error, confirming that human conception is primarily restricted to the six days leading up to and including ovulation.

Limits

The abstract provides no sample sizes, numerical fecundability estimates, or confidence intervals. The analysis relies on statistical assumptions to correct for measurement error rather than direct biological verification of follicular rupture, and it evaluates historical cohorts with disparate populations and study goals (pregnancy avoidance vs. pregnancy seeking).

Cited by