A little under two months ago, the Guttmacher Institute and WHO released first-ever country-level estimates of unintended pregnancy and abortion (see here). The new study analyzed data for 150 countries for the period 2015-2019 and found that:

• Almost half of the 220 million pregnancies globally per year are unintended.
• Six in 10 unintended pregnancies end in an induced abortion (63 million per year).
• Overall, 29% of all pregnancies globally end in an induced abortion.
• Almost 30% of induced abortions occur in countries where abortion is either illegal or heavily restricted. Earlier estimates for 2010-2014 found that 45% of all abortions are unsafe, and around 15% were performed in the least safe conditions, by untrained persons using dangerous and invasive techniques.
• Regional averages mask large disparities within regions for unintended pregnancy and abortion rates.

The Guttmacher/WHO study covers 90% of the 1.9 billion women of reproductive age. Almost all the missing countries (because of lack of data) are from the Western Asia and Northern Africa region, most of them Islamic states or with a dominant Islamic culture. I describe below how I imputed data for most of the missing countries, added data on legal grounds and restrictions regarding abortion and examine global patterns using 12 culture zones which group countries based on history, values and religion.

The following plots give a taste of the global patterns I found. Culture zone and other definitions, data and methods are given in more detail in the rest of this post.

*Note that USA results have been calculated by grouping States into those with and without significant abortion restrictions

The left-hand figure shows that abortion is universally legally available in most of Europe, Canada, Australia and New Zealand, in the Orthodox and Islamic countries of the former Soviet-bloc and in the non-Islamic countries of Asia. It is legally severely restricted in most Islamic countries and sub-Saharan Africa. Abortion rates are substantially lower in the high-income countries of Europe, North America and Australia and New Zealand than in the Asian regions where abortion is unrestricted AND in the countries in all developing regions irrespective of whether abortion is legally restricted or available. Note that USA results have been calculated by grouping States into those with and without significant abortion restrictions.

For 2015-2019, almost half of unintended pregnancies (46%) were aborted in countries where abortion was restricted (often severely) and a little over two-thirds (70%) in countries where abortion is accessible.  In the three regions where around 40-50% of women with unintended pregnancies have restricted access to abortions, overall abortion rates per 1,000 women of reproductive age differ by less than 2 abortions per 1,000 from those in countries (or US states) without restrictions.  

Imputation of data for countries in Western Asia and Northern Africa

The 150 countries with abortion-related data represent 90% of the 1.9 billion women of reproductive age 15-49 years in the period 2015-2019. Of the eight Sustainable Development Goals (SDG) regions used in the study, the coverage of reproductive-age women by countries with data fell below 95% only for Western Asia and Northern Africa, where data coverage was only 3.4%. However, the missing women are almost entirely in one SDG region, Western Asia and Northern Africa

The region includes 24 countries, mostly in North Africa and the Middle East. Data is available only for Azerbaijan, Armenia and Georgia. These are all former Soviet-bloc countries with legal abortion and high rates of termination of unintended pregnancies. In contrast, apart from Israel and Cyprus, the other 19 countries are all either explicitly Islamic states or with a predominant Islamic culture. Seventeen of the 19 either prohibit or severely restrict abortion.

To address the bias inherent in excluding these countries from global statistics, I decided to impute the unintentional pregnancy rates and abortion rates for the Islamic countries in North Africa and the Middle East and also for Afghanistan, Iran and Malaysia, using information on live birth rates for these countries along with the abortion data for 22 Islamic countries in the neighbouring regions Sub-Saharan Africa, South-central Africa and South-eastern Asia. Former Soviet-bloc Islamic countries were not included in this dataset.

Pregnancy rates were regressed against live birth rates for this dataset and the regression model (R-squared 0.91) used to predict pregnancy rates for the missing data. Unintended pregnancies are defined as pregnancies which were unwanted at the time they were conceived or occurred earlier than desired. Explorations of various covariates for predicting the proportion of pregnancies that are unintentional resulted in predicting of missing data using a regression on the log of GDP per capita measured in international (purchasing-power-parity adjusted) dollars and a continuous variable measuring average-religiosity for countries in 2017. Both these variables are significant predictors of the proportion of pregnancies that are unintentional and the regression on the prediction dataset has an R-squared of 0.52. I used previously estimated levels and trends in average religiosity (practicing, non-practicing, non-religious and atheist) for countries using data from the World Values Survey (WVS) and European Values Study (EVS) covering the period 1980 to 2020 (see previous post here). 

The only predictive variable found for the proportion of unintentional pregnancies resulting in induced abortion was the gender inequality index produced by United Nations Development Program. Regression of logit(abortion fraction of unintended pregnancies) against the logit of the 2017 value of the gender inequality index was used to predict missing abortion fractions of unintended pregnancies (R-squared 0.7).

Global and regional abortion rates for 2015-2019

With these missing values imputed for 23 countries, the abortion estimates now cover most countries in most regions, and 98.6% of women of reproductive age globally. Based on this extended dataset, the following table summarizes regional and global patterns of pregnancy and abortion for the period 2015-2019.

The Guttmacher/WHO paper in the British Medical Journal displays unintended pregnancy and abortion rates by country for the 150 countries included in their study. The estimates dataset is available here. It includes 80% and 95% uncertainty intervals.  In this post, I’m interested to see the big picture at regional and global level, and I’ve not done the additional work to estimate uncertainty intervals.  But it is important to keep in mind that there are many problems with data availability, reliability and completeness of reporting, particularly for an often stigmatized and contentious issue such as abortion.  So the broad patterns shown below likely convey a reasonable and evidence-based assessment, but small differences between countries or regions should not be over-interpreted.

Legal restrictions in countries and US States

The legality and acceptability of abortion varies widely across the world, from Europe where it is available without restriction in most countries, to the Middle East and Africa where it is largely illegal.

I’ve taken country-level information on abortion availability and restrictions from Wikipedia, which documents legal grounds on which abortion is permitted, and gestational age limits where relevant, using the following six categories: risk to life, risk to health, rape, fetal impairment, economic or social reasons, on request.  The first four of these grounds cover a fairly small percentage of unintended pregnancies and I have grouped in a single category “restricted or prohibited” countries  which allow abortion only on these grounds. Countries which allow abortions on economic or social grounds or on request (with any gestational age limit set at least 10 weeks or longer) are classified as “unrestricted”. The USA has widely varying legal restrictions on abortions across the 50 states and Washington DC. I have used information on abortion access from Planned Parenthood to classify abortion access as restricted (28 states) and unrestricted (22 states and Washington DC).

Revised culture groups for countries

In earlier posts, I analyzed religious and other values by culture groups for 110 countries using data from the World Values Surveys and the European Values Study. These culture zones were based on the 11 culture zones developed by Welzel (2013). 

For this post, I’ve slightly revised the culture zones to group Canada with Australia and New Zealand in the “Reformed West” zone, and kept the USA as a single separate culture zone. I’ve also revised the categories “Indic East”, “Sinic East”, and Islamic East to group together all countries with a predominantly Islamic culture and values. The revised culture zones are as follows:

Reformed West — Western European societies strongly affected by the Reformation: Denmark, Finland, France, Germany, Iceland, Netherlands, Norway, Sweden, Switzerland, United Kingdom, plus Canada, Australia and New Zealand;

USA — The North American culture zone included only two countries, USA and Canada. Canada’s culture is very similar to the Reformed West, and it is moved there.

Old West — Mostly Catholic parts of Western Europe being core parts of the Roman Empire: Austria, Belgium, Cyprus, Greece, Ireland, Israel, Italy, Luxembourg, Malta, Portugal, Spain;

Returned West — Catholic and Protestant parts of post-communist Europe returning to the EU: Croatia, Czechia, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia;

Orthodox East — Christian Orthodox or Islamic parts of the post-communist world, mostly parts of former USSR;

Indic East — Parts of South and South East Asia under the historic influence
of Indian or Buddhist culture: Bhutan, Cambodia, India, Laos, Malaysia, Myanmar, Nepal, Philippines, Singapore, Sri Lanka, Thailand, Timor-Leste;

Islamic East — Countries with predominantly Islamic religion and culture;

Sinic  East — Parts of East Asia under the historic influence of Chinese culture and/or Buddhism: China, Japan, Mongolia, North Korea, South Korea, Taiwan, Vietnam;

Latin America — Central and South America and the Caribbean;

Sub-Saharan Africa — African countries south of the Sahara.

Oceania — Papua New Guinea and other Pacific Island countries.

Patterns of abortion by culture group

The table below summarizes the abortion situation for countries without legal restrictions and countries with legal restriction in each culture zone. Country-level legal status is used except for the USA where the statistics relate to two groups of states: those with and without legal restrictions on abortion (as defined above). State level estimates are not available for the per cent of unintended pregnancies, or the per cent that are aborted. I have assumed that the national percent of all pregnancies that are unintended (34%) is the same in the two groups of states, and that the % aborted in states without restrictions is the same (38%) as that in the Reformed West where abortion is 100% unrestricted. Since the national level of unintended pregnancies that are aborted is 34%, this implies that the proportion of unintended pregnancies aborted in states with restrictions is 31%.

To graphically illustrate these variations across culture zones, I repeat the two figures included near the beginning of this post:

Globally, 60% of unintended pregnancies ended in abortion in the period 2015-2019. People seek and obtain abortions in all countries, even in those with restrictive abortion laws, where barriers to safe abortion care are high. In fact, over the past three decades, the proportion of unintended pregnancies that end in abortion has increased in countries that have many legal restrictions in place. For 2015-2019, almost half of unintended pregnancies (46%) were aborted in countries where abortion was restricted (often severely) and a little over two-thirds (70%) in countries where abortion is accessible. 

In the three regions where around 40-50% of with unintended pregnancies have restricted access to abortions, overall abortion rates per 1,000 women of reproductive age differ by less than 2 abortions per 1,000 from those in countries (or US states) without restrictions. These are the USA (11 versus 13 per 1,000), the Returned West (11 versus 10 per 1,000) and Latin America (30 versus 31 per 1,000). The Returned West consists of former Soviet-bloc countries that have joined the EU, and the largest of these, Poland, is the only one to have restricted abortion, prohibiting it for fetal impairment, economic or social reasons, or on request.

The figures presented above suggest that the illegalization of abortion will not substantially reduce its incidence.  Over recent decades, most of the changes to the legal grounds for abortion have been in the direction of recognizing women’s rights to reproductive autonomy (recent examples include Ireland, Argentina, Mexico and Columbia). Removal of a right that women have had for 50 years in the USA is an entirely different situation.

The increasing restrictions in the USA are one of the few examples of major reductions in women’s rights occurring outside the Islamic countries where religious extremists have taken control of government. In the case of the USA, these changes are to rights that women have had for half a century and are being driven by an anti-democratic coalition of white nationalists and religious extremists who do not represent the majority views of the population. A recent issue of the Economist identified white evangelicals as the one major group with majority opposition to the legal availability of abortion. A majority of Catholics, mainline Protestants and those with no religious identification think that abortion should be mostly or always legal in the USA, and support is over 75% for Jewish, atheists and non-religious with college education.

The rhetoric of some US extremists, and actions already taken to restrict health insurance coverage for contraceptive use, suggests that further restriction on abortion access may well also be accompanied by further reductions in contraceptive availability. The unintended pregnancy rate may well increase, resulting in an overall increase in numbers of abortions occurring, even if the restrictions reduce the percentage of unintended pregnancies that end in abortion.

In my next post, I will examine differences in maternal mortality across countries and the extent to which they are associated with legal restrictions on abortion.