Evidence quality 3.88/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 4/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 4/5
- D4 Uncertainty
- 3/5
- D5 Scope
- 3/5
- D6 Prose
- 4/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 4/5
● your factors — click this risk ▾ to reveal
- Your factors
≈ As likely as
Perceived
Condoms and the pill are widely described as "99% effective", which is roughly their perfect-use performance, not what happens across a year of ordinary use. In a survey of 500 US women aged 18-45, only 2.2% correctly estimated the typical-use efficacy of both male condoms and combined oral contraceptives, and over two-thirds significantly overestimated the efficacy of each. The perceived risk of a pregnancy on condoms is therefore lower than the measured one, while the specific fear of a condom breaking is probably larger than breakage's actual share of failures.
Rough estimate: Over two-thirds of surveyed women overestimated typical-use condom and pill efficacy, often calling them 99% effective
Actual
~13 in 100 couples pregnant within the first year of typical male-condom use
US women whose contraceptive method was the male condom, 2006-2010 National Survey of Family Growth
Show derivation
Scope is activity_specific_lifetime: the figure is the probability of an unintended pregnancy within the first 12 months of relying on the male condom as the only method (typical use, which includes inconsistent and incorrect use), from Sundaram et al. 2017's analysis of the 2006-2010 NSFG. It is not a US-adult lifetime probability and is not converted into one, because lifetime exposure depends on how many years a person relies on condoms, how often they have sex, whether they switch methods, and fecundity, none of which has a defensible population default. The uncertainty band runs from 10% to 18%. The lower bound is the rounded-down lower limit of a 95% interval around Sundaram's unrounded 12.6% (standard error 1.11, as reported by Bradley et al. 2019: 12.6 +/- 1.96 x 1.11 = 10.4% to 14.8%). The upper bound is not a sampling interval; it reflects measurement and period uncertainty, taking Trussell 2011's typical-use table value and the 1995 NSFG condom figure (both 18%), which rest on older data and different corrections for abortion underreporting. Subgroup and short-window figures (7% for higher-income users, 7% over six cycles in the Walsh 2004 trials) describe variation between couples or shorter exposure, not uncertainty in the average, and are kept out of the band.
Caveats: The headline is a first-year, typical-use rate for couples whose method was the …
The headline is a first-year, typical-use rate for couples whose method was the male condom, measured by the 2006-2010 National Survey of Family Growth; later NSFG cycles may differ, and survey estimates depend on correcting for underreported abortions. It is a per-couple annual probability, not a per-act probability and not a lifetime one. No source used here gives a robust probability of pregnancy from a single act following a condom break; that depends on cycle day and is not estimated. Breakage rates vary widely by setting: 0.4% of condoms in the Walsh 2004 clinical trials versus 5.6% of events among 1990s family-planning clients in Steiner et al. 1994. No source used here splits the 13% into breakage, slippage, incorrect use and non-use; the gap between the 13% typical-use and 2% perfect-use figures is, by definition, what inconsistent and incorrect use add. Failure rates differ by income in the NSFG data (7% at or above 200% of poverty versus 15% at 100-199%); that spread is variation between couples and is not folded into the 10-18% uncertainty band, whose low end comes from the sampling interval around the 12.6% estimate and whose high end reflects older, differently corrected estimates (18%). Estimates for other methods come from different sources (Sundaram's NSFG values versus Trussell's and Contraceptive Technology's tables) and should be compared within a source where possible. CDC data show about 41.6% of US pregnancies in 2019 were unintended, but that share includes pregnancies where no method was used and cannot be read as a contraceptive-failure rate. The perception figures come from a convenience sample of 500 women, not a representative survey. The quoted excerpts were checked on 2026-09-26 against Europe PMC abstracts and the Guttmacher fact sheet; bracketed passages are labelled paraphrases, and Trussell's table values are paraphrased rather than quoted.
Related risks
Other risks on similar themes — for exploring related fears.
Skipping care over ICE fear
How likely is an immigrant adult in the US to avoid needed medical care because of immigration-enforcement fear?
Untreated depression
What fraction of US adults who develop major depression or an anxiety disorder receive no mental health treatment for at least a year?
Drug-resistant infection
What is the risk of developing a serious antibiotic-resistant infection?
Pick challenger
Among US couples relying on the male condom as their only method, about 13 in 100 have a pregnancy within the first year, according to Sundaram and colleagues’ analysis of the 2006-2010 National Survey of Family Growth. With perfect use, meaning a condom every time and put on correctly, the figure is about 2 in 100. The pill sits at about 7 in 100 under typical use, withdrawal at about 20, and IUDs and implants at about 1. With no method at all, Trussell’s 2011 review puts the first-year figure at 85 in 100.
The fear usually takes the form of a condom breaking, but no study located splits the 13% into breakage, slippage, incorrect use and non-use. In the Walsh 2004 clinical trials, 0.4% of condoms broke and 1.1% slipped off, roughly one in 250 and one in 90. Field studies report more: in Steiner’s 1994 study of family-planning clients, breakage occurred in 5.6% of events. The gap between typical use (13%) and perfect use (2%) is, by definition, what inconsistent and incorrect use add over a year, including the times a condom is not used at all. The public picture runs the other way on the overall number. In a 2017 convenience sample of 500 women, over two-thirds overestimated how well condoms and the pill work in typical use, often calling them 99% effective, which is closer to the perfect-use figure than to what couples actually experience.
The 13% is an average across very different couples. In the NSFG data, condom users with incomes at least twice the poverty level had a first-year failure rate of about 7%, versus about 15% among those between one and two times the poverty level. The condoms are the same product, so the gap presumably reflects who uses them and how consistently. The figure is also strictly a first-year rate, and this entry does not convert it into a multi-year or lifetime total: the number of years a person relies on condoms, how often they have sex and whether they switch methods vary too much for a single default to mean anything.
Related tidbits
Condoms are often called 99% effective, yet about 13 in 100 couples relying on them alone have a pregnancy within the first year of typical use; perfect use gets that to about 2. After a vasectomy, the figure is about 15 in 10,000.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] Perspectives on Sexual and Reproductive Health / Sundaram, Vaughan, Kost, Bankole, Finer, Singh & Trussell (Guttmacher Institute) — Contraceptive Failure in the United States: Estimates from the 2006-2010 National Survey of Family Growth
Contraceptive Failure in the United States: Estimates from the 2006-2010 National Survey of Family Growth- Statistic
First-12-month typical-use failure: male condom 13%, withdrawal 20%, IUD and implant 1%; condom failure down from 18% in 1995. Condom failure 7% at incomes at least 200% of poverty vs 15% at 100-199% of poverty.- Excerpt
“[Abstract sentences below checked verbatim against the Europe PMC abstract, 2026-09-26] "Long-acting reversible contraceptives (the IUD and the implant) had the lowest failure rates of all methods (1%), while condoms and withdrawal carried the highest probabilities of failure (13% and 20%, respectively). However, the failure rate for the condom had declined significantly since 1995 (from 18%), as had the failure rate for all hormonal methods combined (from 8% to 6%)." [Paraphrase from abstract, not verbatim: the analysis used 15,728 contraceptive use intervals contributed by 6,683 women; women with incomes at least 200% of the poverty threshold had lower condom failure than those at 100-199% of poverty (7% vs. 15%).] ”
- Source data from
- 2017-02-28
- Accessed
- 2026-09-26
- Calculation
- Headline native = 13/100 couples pregnant within 12 months of typical male-condom use. Normalized = 0.13 (activity-specific, no further transformation). log10(0.13) = -0.886. Multipliers relative to the 13% condom baseline: IUD/implant 1/13 = 0.08; condom users with income at least 200% of poverty 7/13 = 0.54 (against the all-user average, not against the 100-199% group, whose 15% is 1.15x). The pill's 7% is taken from the Guttmacher fact sheet: 7/13 = 0.54. The 1995 figure of 18% sets the upper end of the uncertainty band. Withdrawal's 20% is cited in prose only (20/13 = 1.54, below the 2x factor bar).
- Independence
- Primary NSFG analysis. The Guttmacher fact sheet below summarizes the same Guttmacher-led estimates and is not independent for the 13% figure; Trussell 2011 (a different data basis and editorial synthesis) and the Walsh 2004 clinical trials are the independent cross-checks.
-
[2] Guttmacher Institute — Contraceptive Effectiveness in the United States
Contraceptive Effectiveness in the United States- Statistic
Male condom: typical-use failure 13%, perfect-use failure 2%. Pill: typical-use 7%. Implant and IUD: under 1% for both typical and perfect use.- Excerpt
“[Sentences below checked verbatim against the fact sheet, 2026-09-26] "Contraceptive failure rates are defined as the proportion of women who will become pregnant within the first 12 months after initiating method use." "The male condom has a typical-use failure rate of 13%, and a perfect-use failure rate of 2%." "The pill, ring and patch have typical-use failure rates of 7%, and perfect-use failure rates of less than 1%." [Paraphrase, not verbatim: the implant and the IUD are described as the most effective reversible methods, with failure rates under 1% for both typical and perfect use.] ”
- Source data from
- 2020-04-01
- Accessed
- 2026-09-26
- Calculation
- Supplies the perfect-use condom figure (2%) that the NSFG cannot measure. Perfect-use multiplier = 2/13 = 0.15. The typical-use 13% matches Sundaram 2017. source_date is the fact sheet's stated month (April 2020); the day is set to the 1st.
- Independence
- Guttmacher summary drawing on Sundaram 2017 (typical use) and Contraceptive Technology (perfect use); not independent of source 1 for typical-use figures.
-
[3] Contraception 83(5):397-404 / Trussell J — Contraceptive failure in the United States
Contraceptive failure in the United States- Statistic
Percentage experiencing an unintended pregnancy in the first year: male condom 18% typical use, 2% perfect use; no method 85%- Excerpt
“[Abstract sentences confirmed verbatim via search-engine indexing; journal host unreachable] "This review provides an update of previous estimates of first-year probabilities of contraceptive failure for all methods of contraception available in the United States. Estimates are provided of probabilities of failure during typical use (which includes both incorrect and inconsistent use) and during perfect use (correct and consistent use)." [Paraphrase of the summary table, not verbatim: no method 85%; male condom 18% typical use, 2% perfect use.] ”
- Source data from
- 2011-05-01
- Accessed
- 2026-09-26
- Calculation
- 18% typical-use condom value sets the upper end of the uncertainty band (0.18). No-method 85% gives the no-contraception contrast: 85/13 = 6.5x the condom baseline. Trussell's table was built partly from older NSFG cycles corrected for abortion underreporting, which is why it runs higher than Sundaram's 13%.
- Independence
- Editorial synthesis in Contraception; Trussell is also a co-author of Sundaram 2017, so the two share methodology lineage, but the 2011 table rests on earlier data.
-
[4] Contraception / Walsh, Frezieres et al. — Effectiveness of the male latex condom: combined results for three popular condom brands used as controls in randomized clinical trials
Effectiveness of the male latex condom: combined results for three popular condom brands used as controls in randomized clinical trials- Statistic
Clinical breakage 0.4% and slippage 1.1% over the first five condom uses; six-cycle typical-use pregnancy rate 7.0% (95% CI 5.0-9.0), consistent-use 1.0% (95% CI 0.0-2.1); ~800 couples, 3,526 cycles- Excerpt
“[Abstract sentences confirmed verbatim via search-engine indexing; PubMed and the journal host unreachable] "The combined clinical breakage rate for the first five condom uses was 0.4% for the three latex brands and the combined clinical slippage rate was 1.1%. The combined six-cycle typical-use pregnancy rate for the latex condoms was 7.0% (95% confidence interval 5.0-9.0)." "The combined six-cycle consistent-use pregnancy rate was 1.0% (95% confidence interval 0.0-2.1)." [Paraphrase, not verbatim: data covered 3,526 menstrual cycles from approximately 800 couples using Ramses Sensitol, LifeStyles or Trojan-Enz condoms.] ”
- Source data from
- 2004-11-01
- Accessed
- 2026-09-26
- Calculation
- Per-condom breakage 0.4% (about 1 in 250) and slippage 1.1% (about 1 in 90) under trial conditions. Six-cycle typical-use 7.0% in a motivated trial cohort covers a shorter window than the 12-month headline and is not used in the uncertainty band; it is not annualized here because trial dropout and selection make a straight extrapolation unreliable.
- Independence
- Randomized-trial data, independent of the NSFG survey estimates in sources 1-3.
-
[5] Contraception / Steiner et al. (1994) — Breakage and slippage of condoms in family planning clients
Breakage and slippage of condoms in family planning clients- Statistic
One or more major problems (breakage, slippage, leakage) in 410 of events (10.9%) involving 217 clients (40.2%); breakage in 209 events (5.6%), slippage in 243 (6.5%)- Excerpt
“[Abstract sentence confirmed verbatim via search-engine indexing; PubMed and the journal host unreachable] "One or more major problems, breakage, slippage, leakage or a combination of these, occurred in 410 (10.9%) events, involving 217 (40.2%) clients." [Paraphrase, not verbatim: data came from 540 clients on 3,754 occasions of condom use over one month; breakage occurred in 209 (5.6%) events and slippage in 243 (6.5%).] ”
- Source data from
- 1994-07-01
- Accessed
- 2026-09-26
- Calculation
- Field-use contrast to Walsh 2004: breakage per event 5.6% vs 0.4% in a controlled trial, a roughly 14-fold spread (5.6 / 0.4 = 14). Used in prose and caveats only; not an input to the headline.
- Independence
- Independent 1990s field study in family-planning clients; different population, era and condom stock from Walsh 2004. source_date month approximate.
-
[6] Contraception and Reproductive Medicine / Kakaiya, Lopez & Nelson — Women's perceptions of contraceptive efficacy and safety
Women's perceptions of contraceptive efficacy and safety- Statistic
Of 500 women aged 18-45, 2.2% correctly estimated typical-use efficacy of both combined oral contraceptives and male condoms; over two-thirds significantly overestimated each- Excerpt
“[Abstract sentences confirmed verbatim via search-engine indexing; host unreachable] "A convenience sample of 500 women aged 18-45 years, with education levels that ranged from middle school to postdoctoral level was interviewed." "The efficacy in typical use of both combined oral contraceptives and male condoms was correctly estimated by 2.2%; over two-thirds of women significantly overestimated the efficacy of each of those methods in typical use." "The majority of reproductive aged women surveyed substantially overestimated the efficacy of the two most popular contraceptive methods, often saying that they were 99% effective." ”
- Source data from
- 2017-06-01
- Accessed
- 2026-09-26
- Calculation
- Perception input only. "99% effective" implies a perceived annual failure of about 1% against a measured 13%, a gap of roughly 13-fold for typical use. Not an input to the headline number.
- Independence
- Independent perception survey; unrelated to the NSFG and trial data.
-
[7] Studies in Family Planning 50(1) / Bradley, Polis, Bankole & Croft — Global Contraceptive Failure Rates: Who Is Most at Risk?
Global Contraceptive Failure Rates: Who Is Most at Risk?- Statistic
Cites the Sundaram et al. 2017 NSFG 2006-2010 first-year condom failure estimate as 12.6 (SE 1.11), against a pooled 43-country DHS condom estimate of 8.6- Excerpt
“[Paraphrase from search-engine indexing of the article, not verbatim; PMC and the publisher host unreachable from the authoring environment] In comparing its pooled Demographic and Health Survey estimates with US figures, the article reports the Sundaram et al. (2017) condom failure estimate from 2006-2010 NSFG data as 12.6 with a standard error of 1.11, lower than earlier US rates but higher than its own global condom estimate of 8.6. ”
- Source data from
- 2019-02-01
- Accessed
- 2026-09-26
- Calculation
- 95% interval = 12.6 +/- 1.96 x 1.11 = 12.6 +/- 2.18 = 10.4% to 14.8%. The lower limit, rounded down to 0.10, sets the low end of the uncertainty band. The standard error should be confirmed against Sundaram 2017's own table on verification.
- Independence
- Secondary report of the Sundaram 2017 standard error; not independent of source 1 for the condom figure. Its own DHS estimates are not used here.
-
[8] Centers for Disease Control and Prevention — Unintended Pregnancy | Reproductive Health | CDC
Unintended Pregnancy | Reproductive Health | CDCSee all 2 Likelier entries citing this source →
- Statistic
41.6% of US pregnancies in 2019 were unintended; rate 35.7 per 1,000 females aged 15-44- Excerpt
“"The percentage of U.S. pregnancies that were unintended declined from 43.3% in 2010 to 41.6% in 2019. Unintended pregnancy rates declined by 15%, from 42.1 per 1,000 females aged 15-44 years in 2010 to 35.7 per 1,000 in 2019. Despite this progress, about 4 in 10 pregnancies in the United States are still unintended." ”
- Source data from
- 2024-01-01
- Accessed
- 2026-05-10
- Calculation
- Context only, not an input to the headline. Excerpt reused from the verified citation in the getting-tubal-ligation regret-pair (accessed 2026-05-10); cdc.gov was not reachable from this session. Unintended pregnancies include those with no method in use, so this share cannot be attributed to contraceptive failure alone.
- Independence
- National surveillance-based estimate, independent of the method-specific failure data.