Twin birth rates rose dramatically in the United States from 1980 through the early 2010s, but that long increase has now reversed. The U.S. twin birth rate increased from 18.9 twin births per 1,000 total births in 1980 to a historical high of 33.9 in 2014. By 2024, it had fallen to 30.1 per 1,000 births, an 11% decline from the 2014 peak.
The international picture is more varied. A major global study found that the worldwide twinning rate increased by about one-third between the early 1980s and early 2010s, but countries and regions followed very different paths. International statistics also often measure twin deliveries rather than individual twin babies, so rates from different sources should not automatically be compared as if they use the same denominator.1980 U.S. twin birth rate 18.9 twin births per 1,000 total births 2014 U.S. peak 33.9 twin births per 1,000 total births 2024 U.S. twin birth rate 30.1 twin births per 1,000 total births Change from 2014 to 2024 11% decline in the twin birth rate
U.S. Twin Birth Rate by Year, 1980–2024
The annual U.S. series shows three distinct stages: a long period of rapid growth beginning in the 1980s, slower growth as the rate approached its peak, and a sustained decline after 2014. The rate rose 79% between 1980 and 2014 before falling during the following decade.Key milestones in the U.S. twin birth rate
| Year | Twin births per 1,000 total births | Historical context |
|---|---|---|
| 1980 | 18.9 | Historical baseline for the modern rise |
| 1990 | 22.6 | Rapid increase underway |
| 2000 | 29.3 | Rate substantially above the 1980 level |
| 2010 | 33.1 | Growth had slowed near the eventual peak |
| 2014 | 33.9 | Highest U.S. twin birth rate in the series |
| 2020 | 31.1 | Decline firmly established |
| 2024 | 30.1 | 11% below the 2014 peak |
The full annual series is shown below. The measure is the number of live births in twin deliveries per 1,000 total live births.
| Year | Twin birth rate per 1,000 births |
|---|---|
| 1980 | 18.9 |
| 1981 | 19.3 |
| 1982 | 19.5 |
| 1983 | 19.9 |
| 1984 | 19.9 |
| 1985 | 20.5 |
| 1986 | 21.2 |
| 1987 | 21.5 |
| 1988 | 21.8 |
| 1989 | 22.3 |
| 1990 | 22.6 |
| 1991 | 23.1 |
| 1992 | 23.5 |
| 1993 | 24.1 |
| 1994 | 24.6 |
| 1995 | 24.8 |
| 1996 | 25.9 |
| 1997 | 26.8 |
| 1998 | 28.1 |
| 1999 | 28.9 |
| 2000 | 29.3 |
| 2001 | 30.1 |
| 2002 | 31.1 |
| 2003 | 31.5 |
| 2004 | 32.2 |
| 2005 | 32.2 |
| 2006 | 32.1 |
| 2007 | 32.2 |
| 2008 | 32.6 |
| 2009 | 33.2 |
| 2010 | 33.1 |
| 2011 | 33.2 |
| 2012 | 33.1 |
| 2013 | 33.7 |
| 2014 | 33.9 |
| 2015 | 33.5 |
| 2016 | 33.4 |
| 2017 | 33.3 |
| 2018 | 32.6 |
| 2019 | 32.1 |
| 2020 | 31.1 |
| 2021 | 31.2 |
| 2022 | 31.2 |
| 2023 | 30.7 |
| 2024 | 30.1 |
Data note: Historical rates follow U.S. National Center for Health Statistics natality data. Values for 2014–2024 are confirmed in the 2026 NCHS report A Decade of Decline in Twin Childbearing in the United States, 2014–2024.
1980–2004: Rapid Growth
The modern increase in U.S. twin births became clearly established during the 1980s. The twin birth rate rose from 18.9 per 1,000 births in 1980 to 32.2 by 2004. Earlier NCHS analysis found that increases averaged more than 2% per year from 1980 through 2004, with especially rapid growth during parts of the 1990s.
This was a major change from the preceding historical pattern. U.S. multiple-birth rates had been relatively stable for much of the period from the early twentieth century through the 1970s. The rapid modern increase therefore represented a genuine demographic shift rather than a continuation of a long-standing trend.
2005–2014: Growth Slows and Reaches a Peak
Growth became much slower after the early 2000s. The twin birth rate was 32.2 per 1,000 births in 2005 and fluctuated around 32 to 33 during the following years. It reached 33.7 in 2013 and then a historical high of 33.9 in 2014.
The slowing rate of increase is important because the 2014 peak did not appear suddenly. It followed several years in which the rapid growth seen during earlier decades had already moderated.
The absolute number of twin babies followed a slightly different timeline. CDC data show that the number of twin births peaked earlier, at 138,961 in 2007. The rate continued rising and did not peak until 2014. This illustrates why an absolute birth count and a population birth rate should not be treated as the same statistic.
2014–2024: A Decade of Decline
After reaching 33.9 per 1,000 births in 2014, the U.S. twin birth rate declined to 30.1 in 2024. NCHS describes this as an average decline of about 1% per year and an overall reduction of 11% across the decade.
The number of twin births declined even more sharply, from 135,336 twin babies in 2014 to 109,195 in 2024, a 19% decrease. The number of singleton births also fell during this period, which is one reason the percentage decline in the twin birth count was larger than the decline in the twin birth rate.
The largest single-year decrease during the 2014–2024 period occurred in 2020, when the number of twin births fell 7% and the rate fell 3%. Small increases or pauses occurred in some individual years after that, but the longer-term post-2014 direction remained downward. By 2024, the U.S. twin birth rate was at its lowest level in more than 20 years.
Why Twin Birth Rates Increased
No single factor explains the historical rise in twin births. Two of the most important population-level influences identified in demographic and reproductive-health research are changes in maternal age and the expansion of fertility treatment and medically assisted reproduction.
Maternal Age Shift
Twinning rates vary with maternal age, particularly for fraternal twinning. As the age distribution of mothers shifted toward older ages in the United States and other high-income countries, this demographic change contributed to an increase in the population twin birth rate.
However, maternal age alone does not explain the scale of the modern U.S. increase. An NCHS analysis of the 1980–2009 period estimated that the older age distribution of women giving birth accounted for only about one-third of the rise in the U.S. twinning rate. That leaves a substantial share of the increase associated with other influences.
Fertility Treatment and Assisted Reproduction
The expansion of infertility treatment was another major part of the historical change. Assisted reproductive technology (ART), including in vitro fertilization (IVF), can increase the likelihood of a multiple pregnancy when more than one embryo is transferred. Non-ART fertility treatments that stimulate ovulation can also increase the chance that more than one egg is available for fertilization.
CDC reports examining the historical rise in U.S. twin births have identified the increased use of infertility treatments, including both ART and non-ART treatment, as an important likely contributor. International research uses the broader term medically assisted reproduction (MAR) and similarly identifies its spread as a major driver of rising twinning rates in many higher-income countries.
The relative contribution of maternal age and fertility treatment varies across countries and time periods. A national increase in twin births should therefore not be attributed to one mechanism without appropriate evidence.
Why Twin Birth Rates Have Declined Since the Peak
The recent decline does not mean reproductive medicine has become less important. Instead, reproductive practice itself has changed. One important development has been greater emphasis on reducing multiple-fetus pregnancies during fertility treatment.
Changes in Embryo-Transfer Practice
Historically, transferring more than one embryo during IVF could improve the chance that at least one embryo implanted, but it also increased the chance of twins or higher-order multiples. Modern practice has increasingly moved toward transferring a single embryo when clinically appropriate.
CDC ART surveillance shows how substantial this change has been in the United States. The percentage of embryo-transfer cycles in which a single embryo was transferred increased from 30.8% in 2013 to 85.9% in 2022.
This shift overlaps with the period in which the national twin birth rate moved downward. It is reasonable to view changing ART practice as part of the context for the decline, but the national trend should not be presented as proof that single-embryo transfer alone caused the entire decrease. Population twin rates also reflect maternal age patterns, non-ART fertility treatment, total birth patterns, and other demographic changes.
Reducing Multiple Pregnancy Risk
One reason reproductive medicine has moved toward single-embryo transfer is that multiple gestation carries greater maternal and infant health risks than singleton pregnancy. CDC describes single-embryo transfer as a strategy for reducing multiple-fetus pregnancies and associated outcomes such as prematurity and low birth weight.
The same technology that contributed to the historical increase in multiple births can therefore have a different population effect when clinical practice changes. This helps explain why the relationship between fertility treatment and twin birth trends is not simply “more IVF means more twins.”
How Twin Birth Rates Are Measured
Twin statistics can look contradictory when two sources use different definitions. The most important question is not only the reported number, but also what was counted in the numerator and denominator.
| Measure | Numerator | Denominator | Common use |
|---|---|---|---|
| Twin births per 1,000 births | Individual babies born in twin deliveries | All births | CDC/NCHS U.S. reporting |
| Twin deliveries per 1,000 deliveries | Deliveries producing twins | All deliveries | HMBD and international demographic research |
Twin Babies per 1,000 Births
The CDC twin birth rate counts individual live-born babies from twin deliveries and expresses that number per 1,000 total births. A U.S. rate of 30.1 in 2024 therefore means that about 30 of every 1,000 live-born babies were born as twins under this measure.
This is a birth-level statistic. It does not mean that 30 of every 1,000 pregnancies or deliveries resulted in twins.
Twin Deliveries per 1,000 Deliveries
The Human Multiple Births Database and much of the international demographic literature instead use a delivery-level twinning rate. One delivery resulting in a pair of twins counts as one twin delivery, not two twin births.
A simple hypothetical example shows why the two numbers can look very different. Suppose there are 1,000 deliveries, including 15 twin deliveries and 985 singleton deliveries. Those deliveries produce 1,015 babies: 30 twin babies plus 985 singletons.
- The delivery-level twinning rate is 15 twin deliveries per 1,000 deliveries.
- The birth-level twin rate is about 29.6 twin babies per 1,000 babies born.
Both figures can describe the same underlying population. They simply use different units and denominators. A rate near 15 in an international twin-delivery dataset is therefore not automatically inconsistent with a birth-level rate near 30.
Twin Birth Rate vs Number of Twin Births
An absolute count answers, “How many twin babies were born?” A rate answers, “How common were twin births relative to all births?” These statistics can reach their highs in different years because the total number of births in the population also changes.
In the United States, the number of twin births peaked in 2007, while the twin birth rate peaked seven years later in 2014. From 2014 to 2024, the twin birth count fell 19%, while the rate fell 11%.
For trend analysis, both measures are useful, but they answer different questions.
Global Twin Birth Trends
The historical increase in twins was not limited to the United States. A large peer-reviewed analysis covering 165 countries estimated that the worldwide twinning rate increased from 9.1 twin deliveries per 1,000 deliveries in 1980–1985 to 12.0 in 2010–2015, an increase of about one-third.
The study estimated that more than 1.6 million twin pairs were being born each year by 2010–2015. It linked much of the increase in wealthier regions to the expansion of medically assisted reproduction, with delayed childbearing also contributing.Regional twinning rates in standardized study periods
| Region | 1980–1985 | 2010–2015 |
|---|---|---|
| Africa | 16.5 | 17.1 |
| Asia | 7.0 | 9.2 |
| Europe | 9.1 | 14.4 |
| North America | 9.9 | 16.9 |
| Oceania | 10.1 | 14.8 |
| South America | 8.7 | 9.3 |
| World | 9.1 | 12.0 |
Unit: twin deliveries per 1,000 total deliveries. These figures come from standardized regional comparisons in the 2021 global study and should not be compared directly with the U.S. CDC birth-level rate without accounting for the different denominator.
The International Rise in Twinning
The global rise was highly uneven. North America, Europe, Oceania, and parts of Asia experienced large increases. The global study found increases greater than 10% in 74 of the 112 countries for which both historical periods could be compared.
The geographic starting point also differed substantially. African countries historically had high natural twinning rates, while much of East Asia and South America had lower rates. As medically assisted reproduction became more widely used, some regions that had previously had moderate or low rates experienced much faster increases.
Countries Have Followed Different Paths
A global average can hide very different national trajectories. The international study highlighted rapid historical increases in countries including Greece, Denmark, and South Korea, while the overall increase across Africa was comparatively small during the standardized study periods.
More recent data also show that the direction is not permanently upward. In some countries, twin rates have stabilized or declined as fertility-treatment practice has changed, including a shift from transferring multiple embryos toward single-embryo transfer.
For that reason, the statement “twin births are increasing worldwide” is too broad when discussing the current period. The historical global increase is well documented, but the latest direction depends on the country, measurement method, and years being compared.
Why Twin Birth Rates Cannot Always Be Compared Directly
International twin statistics are useful only when their definitions and coverage are understood. Two numbers can differ because populations genuinely have different twinning rates, but they can also differ because the underlying data were collected or calculated differently.
Different Denominators
The first check should always be the denominator. A statistic based on twin babies per 1,000 births cannot be directly ranked against one based on twin deliveries per 1,000 deliveries without conversion and appropriate assumptions.
Some national statistical systems also publish broader multiple-birth or multiple-maternity measures that can include triplets and higher-order multiples. Those measures answer a different question again.
Stillbirth Coverage and Registration Rules
International databases draw on national vital statistics, civil registration systems, surveys, medical registrations, and published research. Definitions of a registrable birth or delivery can differ across countries and historical periods.
Some series include stillbirths in the delivery denominator or numerator while others are restricted to live births. Registration completeness can also vary. These differences may be small in some comparisons but material in others, especially when countries have very similar reported rates.
Different Latest Years and Data Revisions
Countries do not all release final birth statistics on the same schedule. A table labelled “latest” may therefore compare one country’s 2022 value with another country’s 2020 or 2024 value.
National agencies may also revise historical figures as registration improves or datasets are updated. For rigorous comparisons, check the metric definition, source, reference year, coverage, and revision status before interpreting a difference as a genuine demographic gap.
Comparison rule: Before comparing two twin birth rates, confirm that they use compatible numerators, denominators, birth-registration rules, and reference periods.
How Recent U.S. Trends Differ Across Groups
The national decline since 2014 has not been uniform across all maternal-age and race or ethnicity groups. Recent NCHS data show substantial differences beneath the overall U.S. average.
Maternal Age
From 2014 to 2024, the largest declines occurred among mothers age 30 and older. The twin birth rate fell:
- 18% for mothers ages 30–34, from 40.3 to 33.2 per 1,000 births.
- 27% for mothers ages 35–39, from 48.6 to 35.4.
- 46% for mothers age 40 and older, from 66.0 to 35.6.
- 5% for mothers ages 25–29, from 30.5 to 28.9.
Changes among mothers ages 20–24 and those younger than 20 were not statistically significant over the same period.
These figures illustrate why a national twin birth rate should not be interpreted as though every maternal-age group has the same underlying rate or follows the same trend.
Race and Ethnicity
Recent trends also differed among the three largest race and Hispanic-origin groups reported by NCHS. Between 2014 and 2024, twin birth rates declined 15% among White non-Hispanic mothers and 3% among Hispanic mothers, while the rate increased 5% among Black non-Hispanic mothers.
| Group | 2014 rate | 2024 rate | Change |
|---|---|---|---|
| White non-Hispanic | 36.7 | 31.3 | -15% |
| Black non-Hispanic | 40.0 | 42.0 | +5% |
| Hispanic | 24.1 | 23.4 | -3% |
These are descriptive population statistics. The differences should not be used on their own to infer biological, social, medical, or treatment-related causes.
What Population Twin Birth Rates Mean for Your Personal Odds
A population twin birth rate is not the same thing as an individual’s probability of having twins.
For example, the 2024 U.S. figure of 30.1 twin births per 1,000 total births describes how many individual babies in the national birth population were born in twin deliveries. It is not a statement that every pregnancy has a 3.01% chance of producing twins.
Personal twin probability can vary with characteristics such as maternal age, reproductive history, fertility treatment, and other factors. Population statistics are useful for understanding demographic patterns, but they should not be converted directly into an individual forecast.
Data Sources and Methodology
This page prioritizes primary government statistics and authoritative demographic research. U.S. and international rates are kept conceptually separate when they use different measurement systems.
U.S. Data
U.S. twin birth statistics come from the National Center for Health Statistics (NCHS), part of the Centers for Disease Control and Prevention (CDC). NCHS natality statistics are based on birth-certificate information collected through the National Vital Statistics System (NVSS).
The historical U.S. series uses the NCHS definition of the twin birth rate: the number of live births in twin deliveries per 1,000 total live births. Historical values are documented in NCHS natality files and reports, with 2014–2024 values updated from the June 2026 Health E-Stat.
Key U.S. sources include:
- CDC/NCHS: A Decade of Decline in Twin Childbearing in the United States, 2014–2024
- CDC/NCHS: Three Decades of Twin Births in the United States, 1980–2009
- CDC FastStats: Multiple Births
- CDC/NCHS Vital Statistics Online Data Portal
International Data
International context is based primarily on the Human Multiple Births Database (HMBD) and peer-reviewed demographic research using twin deliveries per 1,000 total deliveries.
The standardized regional comparison on this page uses the 1980–1985 and 2010–2015 periods from the study Twin Peaks: More Twinning in Humans Than Ever Before. That research assembled information for 165 countries covering more than 99% of the global population for the later study period.
- Human Multiple Births Database
- Human Reproduction: Twin Peaks: More Twinning in Humans Than Ever Before
Definitions and Limitations
Twin birth rate: On the U.S. sections of this page, this means individual live-born babies from twin deliveries per 1,000 total live births.
Twinning rate: In the international demographic sections, this generally means twin deliveries per 1,000 total deliveries.
Twin birth count: The absolute number of babies born in twin deliveries. It is not interchangeable with a rate.
Percentage changes: Published percentage changes are reported as given by the cited source. Where this page describes patterns from table values, figures are interpreted using the stated units and time periods.
International comparison: Country data can differ in their inclusion of stillbirths, registration rules, coverage, source type, and latest available year. A difference between two reported rates should therefore not automatically be interpreted as a true difference in the biological frequency of twinning.
Causality: Historical changes in maternal age, fertility treatment, and embryo-transfer practice provide important explanatory context, but descriptive population trends alone do not prove that one factor caused a particular national increase or decrease.
Update Policy
This resource should be reviewed when NCHS publishes new final U.S. natality data and when major international twin-birth datasets or peer-reviewed comparative studies are updated. The current U.S. series runs through 2024, the latest final year available in the June 2026 NCHS reporting used here.
International series are updated on different schedules. For that reason, the main global comparison on this page uses standardized historical study periods rather than combining different countries’ latest years into one supposedly equivalent comparison.
Frequently Asked Questions
Does a U.S. twin birth rate of 30.1 per 1,000 mean 3.01% of pregnancies produce twins?
No. The CDC measure counts individual babies born in twin deliveries relative to all live-born babies. It is not the percentage of pregnancies or deliveries that result in twins. A delivery-level twinning rate uses a different numerator and denominator.
Why can one source report a twin rate near 30 while another reports a rate near 15?
The sources may be measuring different things. A birth-level statistic can count both babies in a twin pair, while a delivery-level statistic counts the pair as one twin delivery. Before comparing rates, check whether the denominator is total births or total deliveries.
Are twin birth rates still increasing?
Not everywhere. The long-term global increase since the 1980s is well documented, but the recent U.S. trend is downward: the rate fell from 33.9 in 2014 to 30.1 in 2024. Some other countries have also experienced stabilization or declines, while patterns elsewhere differ.
Why do international twin datasets show different latest years for different countries?
National birth-registration systems publish and revise data on different schedules. International databases may therefore contain a newer observation for one country than another. Reference year should always be checked before ranking or comparing countries.
Can the national twin birth rate be used as my personal chance of having twins?
No. A national rate describes births across an entire population. Individual probability depends on personal and reproductive factors and should not be inferred directly from the population birth rate.
What year did twin births peak in the United States?
It depends on which statistic is meant. The U.S. twin birth rate peaked at 33.9 per 1,000 births in 2014. The absolute number of twin births peaked earlier, in 2007. This is why rate and count should be reported separately.