Costs of Induced Abortion and Cost-Effectiveness of Universal Access to Modern Contraceptives in Uganda
Published on: Mar 4, 2016
Transcripts - Costs of Induced Abortion and Cost-Effectiveness of Universal Access to Modern Contraceptives in Uganda
With a population of nearly 30 million and an annual population growth rate of 3.2 percent, Uganda is the third fastest-growing country in the world.1 Recent Demographic and Health Surveys indicate that only 31 percent of Ugandan women of reproductive age who want to use contraceptives report that they are indeed using a modern effective method. Over 37 percent of women want to postpone or limit childbearing but are not currently using modern contraception—these women have an unmet need for family planning.2 To respond to the high level of unmet need, the Ugandan government has begun to include family planning in its health program and has acknowledged that a high level of unmet need for family planning may negatively affect women’s health and overall well-being (see box, page 2).
This research brief highlights findings from two recent studies led by Joseph Babigumira and researchers at the University of Washington and in Uganda. Both studies use health economics methods to:
••Investigate the economic consequences of not responding to unmet need for contraception.
••Inform policymakers about the benefits of increasing family planning coverage.
The first study, published in 2011, assesses the costs and the economic burden associated with induced abortions in Uganda; the second study, completed in 2012, examines the potential costs and health benefits to increasing access to modern contraceptives. The two studies agree that providing greater access to contraception in Uganda may be highly cost effective by alleviating unmet need for family planning services, reducing the incidence of induced abortions and abortion-related complications, and promoting overall reproductive health and well-being.
The Costs of Induced Abortion
Abortions are illegal in Uganda under all circumstances except to save the life of the mother. Despite the legal barriers and health risks, the demand for induced abortions in Uganda is high. In 2009, there were an estimated 362,000 cases of induced abortion in Uganda.3 Moreover, the high number of unintended pregnancies, a result of the very high unmet need for contraception, keeps the demand high for abortion services.
Abortion-related complications are associated with numerous adverse health consequences and exert a substantial cost burden on the Ugandan health care system. The 2011 study aimed to:
••Identify the health and economic impact of induced abortions and abortion-related complications.
••Estimate the cost burden associated with induced abortions.4
To perform the analysis, the researchers developed a decision tree model that identified the possible health outcomes and costs associated with an induced abortion. The model shows the consequences of an abortion in several stages:
••Women seek abortion services from trained practitioners or from untrained providers.
••Women who receive abortion procedures from the different providers are further grouped by whether the induced abortion succeeds or fails.
••Women who have successful induced abortions are divided by whether or not they develop post- abortion complications.
COSTS OF INDUCED ABORTION AND COST-EFFECTIVENESS OF UNIVERSAL
ACCESS TO MODERN CONTRACEPTIVES IN UGANDA
BY MAHESH KARRA AND JAMES N. GRIBBLE
The cumulative national expenditure on induced abortion in 2009 was more than 4% of Uganda’s annual health care expenditure.
The average number of
children that Ugandan women have over their reproductive lives.
Over 37% of Ugandan women want to postpone or limit childbearing but are not currently using modern contraception.
2 www.prb.org COSTS OF INDUCED ABORTION
••Women who develop complications are further classified by
whether they require outpatient care or hospital care, and
whether they have access to such care and services.
••Women who need hospital treatment following abortion
complications but are unable to access it are divided into
those who die at home and those who belatedly seek
hospital care, which is commonly reported in Uganda.
•• In the case of abortion failure by a practitioner, the model
assumes that women carry the pregnancy and face
Three types of costs incurred by women and their families were
obtained from published data sources. Probabilities were calcu-lated
and assigned to each event at every stage:
•• Direct medical costs, which include costs for medical
supplies, tests and screenings, and any out-of-pocket
•• Direct nonmedical costs, such as transportation costs,
while seeking health care.
•• Indirect costs, which encompass loss of time and
productivity while seeking abortion services and getting
treatment for complications.
The study also estimated the total direct cost incurred by women
and their families (the direct medical and nonmedical costs
associated with procuring an abortion), and the total government
costs associated with treating abortion-related complications.
In addition, the societal costs of induced abortion are estimated
by aggregating direct medical and nonmedical costs incurred
by women and their families, indirect costs associated with time
and productivity loss, and total government costs.
An induced abortion is associated with $177 in societal costs,
four times higher than the level of per capita health expenditure
in Uganda. Moreover, 52 percent of the total societal cost can be
attributed to indirect costs and costs associated with productiv-ity
loss, while the remaining 48 percent is associated with the
direct costs for providing health care. Women and their families
bear over 83 percent of the total direct cost burden associated
with induced abortion, whereas the government, which is the
primary health care provider in Uganda, incurs only 17 percent.
Based on the estimated number of induced abortions that were
performed in 2009, the researchers estimated that the cumula-tive
national expenditure on induced abortion in 2009 was $64
million in societal costs—more than 4 percent of Uganda’s total
annual health care expenditure of approximately $1.5 billion.
Given the increased demand for induced abortion because
of the high number of unintended pregnancies, the analysis
emphasizes the need to improve contraceptive coverage. The
substantial costs associated with unsafely induced abortions in
Unmet Need, Unintended
Pregnancy, and Induced Abortion
In Uganda, low levels of contraceptive use and widespread
unmet need are associated with large numbers of unintended
pregnancies and unplanned births. Studies show that the num-ber
of unintended pregnancies and births in Uganda remains
high—almost half of all births in 2006 were unplanned.1
Sexually active women who do not use contraception and do
not want to have more children are at high risk of having an
abortion should they become pregnant. Ugandan law permits
an abortion only when a pregnancy endangers a woman’s
life; under these restricted conditions, legal abortions are
rarely approved.2 Nevertheless, induced abortions in Uganda
are common and usually result in serious health complica-tions
because procedures are often performed by untrained
personnel in unsafe, unhygienic environments. Furthermore,
treatments for abortion-related complications are costly and
consume scarce medical and health resources, and many
women who receive induced abortions suffer from short-term
and long-term health consequences.
In Uganda, two of every five pregnancies end in induced abor-tion,
and one in two pregnancies are unintended. These data
are consistent with the high level of unmet need for contracep-tion
in Uganda.3 A significant part of this unmet need may be
attributed to a lack of access to contraceptive methods. Two of
every three Ugandan women who want family planning do not
have access to modern methods of contraception, in spite of
government efforts to increase coverage. Moreover, the range
of methods provided at health facilities, particularly at the
district and local levels, is limited. Although increasing access
to contraceptive methods may be beneficial to women and
their families, Uganda’s publicly provided health care system
has many competing health and development needs because
of severe budget and resource constraints. Consequently,
many public health interventions, including efforts to improve
contraception coverage or to provide safe and legal abortions
where permitted, are compromised.
1 Uganda Bureau of Statistics (UBOS) and Macro International Inc.,
Uganda Demographic and Health Survey 2006 (Calverton, MD:
UBOS and Macro International Inc., 2007).
2 Susheela Singh et al., Unintended Pregnancy and Induced Abortion
in Uganda: Causes and Consequences (New York: Guttmacher
3 Florence M. Mirembe, “A Situation Analysis of Induced Abortions
in Uganda,” African Journal of Fertility, Sexuality, and Reproductive
Health 1, no. 1 (1996): 79-80.
COSTS OF INDUCED ABORTION www.prb.org 3
Uganda highlight the need to make family planning information
and services more available to Ugandan women and couples
who want them; to reduce the number of unsafe abortions;
and to provide safe and legal abortion and post-abortion care
services as permitted under Ugandan law.
Increased Access to Contraception:
A Cost-Effectiveness Analysis
A second study examined two closely related questions: How
would increasing contraceptive coverage affect women’s health
and well-being in Uganda? Would such an intervention be a
good use of scarce health resources? To answer these ques-tions,
Babigumira and colleagues compared a hypothetical
new program, which would provide universal access to modern
contraception, to the current program based on 2006 DHS data,
where contraceptive availability is poor.5 The study:
•• Defines the analytic approach used in a new hypothetical
•• Examines changes in sexual activity, contraceptive use,
and pregnancy over a woman’s life course under the new
•• Calculates the costs and benefits associated with key health
indicators under the new program.
•• Compares the projected costs and benefits associated with
the hypothetical new program to those associated with the
The study used 2006 DHS data to identify the distribution of
Ugandan women who were either using a modern method, a
traditional method, or not using a method under the current
program but did not want to become pregnant for two years
or more. Assuming that these women all have an interest in
using modern contraception, the researchers applied the same
contraceptive method mix observed by women in the 2006
survey to the expanded group of women. Consequently, if the
new program had been implemented in 2006, an estimated 3.2
million sexually active Ugandan women of reproductive age who
desired contraception would have been given access to and
would have used modern methods. Through maintaining the
same method mix, the new program simulates an environment in
which all women who want to avoid pregnancy have access to
contraception and unmet need for family planning is eliminated.
Table 1 illustrates the contraceptive method mix used by
women who desire a contraceptive method under the cur-rent
program (CCP) and under the hypothetical new program
Contraceptive Use Under CCP and NCP
CURRENT PROGRAM (CCP) NEW PROGRAM (NCP)
NUMBER % NUMBER %
All Sexually Active Women Who Desire Contraception 3,200,000 100.0 3,200,000 100.0
No contraception 1,952,000 61.0 0 0
Any method 1,248,000 39.0 3,200,000 100.0
Any modern method 992,000 31.0 3,200,000 100.0
Female sterilization 108,800 3.4 352,000 11.0
Male sterilization 6,400 0.2 19,200 0.6
Pill 147,200 4.6 473,600 14.8
Intrauterine device (IUD) 6,400 0.2 19,200 0.6
Injectable 496,000 15.5 1,600,000 50.0
Implants 19,200 0.6 60,800 1.9
Male condom 204,800 6.4 659,200 20.6
Any traditional method 256,000 8.0 0 0
Rhythm 124,800 3.9 0 0
Withdrawal 86,400 2.7 0 0
Folk methods 44,800 1.4 0 0
Source: Joseph B. Babigumira et al., “Potential Cost-Effectiveness of Universal Access to Modern Contraceptives in Uganda,” PloS one 7, no. 2 (2012): 1-9.
4 www.prb.org COSTS OF INDUCED ABORTION
Using the new distribution as a starting point, the study devel-oped
a Markov cohort model to simulate the impact of the new
program on the reproductive health experiences of Ugandan
women over the course of their lives. The simulation follows a
hypothetical cohort of women who move between seven health
states starting at age 15 (approximately the median age of
sexual debut in Uganda) and ending at age 49 (assumed to be
the end of a woman’s reproductive lifetime).6
The two programs are compared across three measures:
•• Total costs (direct and indirect costs, medical and
•• Life expectancy (measured in life years), and disability-adjusted
•• Cost effectiveness, defined as cost per life-year saved and
cost per disability-adjusted life-year (DALY) averted. (One
DALY is equivalent to one year of healthy life lost.)
The analysis also compares the two programs across indicators
of health status, including indicators associated with pregnancy
and pregnancy-related complications; and measures of neona-tal,
infant, and child mortality.
In the model, only two health states—pregnancy and modern
contraception—incur any costs. All other states in the model—
sexual inactivity, intentionally not using contraception, using
traditional family planning methods, unintentionally not using
contraception, and death—are assumed to have no costs. For
the two programs, costs were calculated for pregnancy and
modern contraception use, accounting for expenditures incurred
by the government (Ugandan Ministry of Health) as well as soci-etal
costs. Governmental costs include direct and indirect medi-cal
costs incurred by the Ministry of Health, the primary health
care provider in Uganda. Societal costs include governmental
costs, any direct nonmedical expenditure by patients (such as
transportation costs), and costs resulting from lost productivity.
The study predicted that for a cohort of 100,000 15-year-old
Ugandan women, the new hypothetical program would result
in reduced costs and more favorable health outcomes, thereby
outperforming the current program.
Given the expected expansion in contraceptive services under
the new program, program costs would be higher than under
the current program. Societal costs under the new program
would be higher by $225 per woman and governmental costs
would be higher by $52 per woman. However, medical costs
per woman under the new program would be lower from both
societal ($263) and governmental ($101) perspectives. The
lower medical costs under the new program would offset the
higher program costs, making the new program less costly than
the current program. In fact, the analysis shows that increasing
access to contraception under the new program would lower
societal costs by $38 per woman, lower governmental costs by
$48 per woman, and save close to $4 million.
The analysis also highlights the positive relationship between
increased contraceptive coverage and improved maternal and
child health outcomes. Under the new program, the total fertility
rate would decrease from 6.9 children per woman to 5.8 chil-dren
per woman. In addition, women would have fewer abor-tions,
miscarriages, and stillbirths. Child and infant mortality rates
would also decrease under the new program.
When assessing life expectancy indicators, the study suggests
that women would live longer under the new program (27.38
DALYs averted per woman) than under the current program
(27.01 DALYs averted per woman).
The estimates from the analysis indicate that the new program is
both less expensive and more effective than the current program
(see Table 2). Under the new program, increasing contracep-tive
use would result in more favorable health outcomes (higher
COST PER WOMAN
COST PER WOMAN
Societal Perspective Current Program $1,987 27.01
New Program $1,949 $38 SAVED 27.38 0.37
Governmental Perspective Current Program $684 27.01
New Program $636 $48 SAVED 27.38 0.37
Cost-Effectiveness Analysis and Program Comparison
Source: Joseph B. Babigumira et al., “Potential Cost-Effectiveness of Universal Access to Modern Contraceptives in Uganda,” PloS one 7, no. 2 (2012): 1-9.
202 483 1100 PHONE
202 328 3937 FAX
1875 Connecticut Ave., NW
Washington, DC 20009 USA
POPULATION REFERENCE BUREAU
POPULATION REFERENCE BUREAU
The Population Reference Bureau INFORMS people around the world about population, health, and the environment, and EMPOWERS them to use that information to ADVANCE the well-being of current and future generations.
DALYs averted per woman) and lower costs than under the current program. In health economics terms, the new program is said to “dominate” the old program and such a program is the best possible use of scarce resources. On the other hand, competing health interventions, such as treatment for measles or home-based antiretroviral therapy, while worth spending resources on, would not provide the same level of benefits from a societal and from a governmental perspective.
Many studies have shown that contraception has a positive impact on maternal and child health outcomes. In a country like Uganda, which has one of the highest maternal mortality rates in the world at 435 maternal deaths per 100,000 live births, limiting unintended pregnancies can decrease women’s risks from unsafely performed abortions, miscarriages, and complicated pregnancies.7 The benefits of universal access would be critical in alleviating abortion-related complications and allowing couples to have the number of children they desire.
The health economics studies highlighted in this brief indicate that, in addition to the health benefits associated with universal access, increasing contraceptive coverage is an effective use of scarce health care resources. Improving access to family planning is essential to reducing unintended pregnancies and induced abortions. The researchers encourage policymakers to focus on solutions that:
••Improve contraceptive supply, especially at lower-level health care facilities.
••Broaden the range of contraceptive methods offered at hospitals and clinics.
••Expand contraceptive service delivery strategies that serve people in remote areas.
••Train more health care providers in family planning counseling and service delivery, especially for long-acting and permanent methods (such as intrauterine devices, implants, and sterilization).
••Improve public knowledge of contraceptive methods.
At a minimum, the results suggest that family planning and reproductive health care should be a high priority in Uganda and should be covered under the national health care scheme. The benefits from such high-impact policy efforts can contribute to Uganda’s health, social, and economic development.
This brief was prepared by Mahesh Karra, policy analyst in International Programs, and James N. Gribble, vice president of International Programs, at PRB. This publication is made possible by the generous support of the William and Flora Hewlett Foundation, as part of the foundation’s Population and Poverty Research Network (PopPov). The authors wish to thank Joseph Babigumira of the University of Washington and his colleagues for their technical contributions to this brief.
© 2012 Population Reference Bureau. All rights reserved.
1 Shane Khan et al., Unmet Need and the Demand for Family Planning in Uganda: Further Analysis of the Uganda Demographic and Health Surveys, 1995-2006 (Calverton, MD: Macro International Inc., 2008).
2 Uganda Bureau of Statistics (UBOS) and Macro International Inc., Uganda Demographic and Health Survey 2006 (Calverton, MD: UBOS and Macro International Inc., 2007).
3 Michael Vlassoff et al., “Benefits of Meeting the Contraceptive Needs of Ugandan Women,” In Brief no. 4 (Washington, DC: Guttmacher Institute, 2009); and Michael Vlassoff et al., “Estimates of Health Care System Costs of Unsafe Abortion in Africa and Latin America,” International Perspectives on Sexual and Reproductive Health 35, no. 3 (2009): 114-21.
4 Joseph B. Babigumira et al., “Estimating the Costs of Induced Abortion in Uganda: A Model-Based Analysis,” BMC Public Health 11, no. 1 (2011): 904.
5 Joseph B. Babigumira et al., “Potential Cost-Effectiveness of Universal Access to Modern Contraceptives in Uganda,” PloS one 7, no. 2 (2012): 1-9.
6 The seven stages of the Markov model are: not sexually active; intentional noncontraception; modern contraception; traditional contraception; unintentional noncontraception; pregnant; and death.
7 UBOS and Macro International Inc., Uganda Demographic and Health Survey 2006.