Dunson · Obstetrics and gynecology 2004 · prospective cohort study · n=782

Increased infertility with age in men and women.

Cited 521 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective observational cohort study evaluating age-related fecundability and time-to-conception.

PubMed 14704244 · doi:10.1097/01.AOG.0000100153.24061.45 · record verified 2026-08-26

What was done

A prospective fecundability cohort study tracked 782 outwardly healthy couples recruited from seven European natural family planning centers. Eligible women were aged 18-40 years. Daily intercourse records were used to adjust for timing and frequency of intercourse when estimating the per-menstrual-cycle probability of conception. Menstrual cycles required to achieve clinical pregnancy, alongside probabilities of sterility and 12-cycle infertility, were modeled across female and male age groups.

What was found

Sterility was estimated at approximately 1% and did not change with age. Estimated 12-cycle infertility rates by female age were 8% for ages 19-26 years, 13-14% for ages 27-34 years, and 18% for ages 35-39 years. Male age was an important factor starting in the late 30s, with failure to conceive within 12 cycles increasing from an estimated 18% to 28% between male ages 35 and 40 years. Among couples unable to conceive in 12 cycles, an estimated 43-63% (varying by age) were projected to conceive with an additional 12 cycles of trying.

Why it matters

This study shows that age-related infertility reflects slower conception rates rather than absolute sterility, indicating that many older couples meeting clinical infertility criteria will still conceive spontaneously within a second year.

Limits

The cohort was restricted to European natural family planning users and women aged 18-40, limiting generalizability to broader populations or women over 40. Estimates rely on statistical modeling of fecundability distributions rather than direct long-term follow-up beyond standard observation windows, and specific biological confounders (e.g., semen parameters, subclinical pathology) were not reported in the abstract.

Cited by