Cost Savings From Emergency Contraceptive
Pills in Canada
JAMES TRUSSELL, PhD, ELLEN WIEBE, MD, TARA SHOCHET, MPH, AND
ÉDITH GUILBERT, MD, MSc
Objective: To estimate cost savings from emergency contra-
ceptive pills in Canada.
Methods: We modeled cost savings when a single emer-
gency contraceptive treatment was provided after unpro-
tected intercourse and when women were provided emer-
gency contraceptive pills in advance.
Results: Each dollar spent on a single treatment saved
$1.19 –$2.35 (in Canadian currency), depending on the regi-
men and on assumptions about savings from costs avoided
by preventing mistimed births. The dedicated products
Preven (Shire Canada, Inc., Oakville, Ontario) and Plan B
(Paladin Labs, Inc., Montreal) were cost-saving even under
the least favorable assumption that mistimed births pre-
vented today occur 2 years later. Each dollar spent on
advance provision of Preven saved $1.24 –$12.23, depending
on the regular contraception method, on how consistently
emergency contraception was used when needed, and on
whether mistimed births were averted forever or simply
delayed. Plan B was almost always cost-saving, although less
so.
Conclusion: Emergency contraception was cost-saving
whether provided when the emergency occurred or in ad-
vance to be used as needed. More extensive use of emer-
gency contraception could save considerable medical costs
by reducing unintended pregnancies. (Obstet Gynecol 2001;
97:789 –93. © 2001 by The American College of Obstetri-
cians and Gynecologists.)
Emergency contraceptive pills, which prevent preg-
nancy after unprotected sexual intercourse, were found
to be cost-saving in the United States.1 In this study, we
estimate cost savings generated by use of emergency
contraception in Canada by comparing costs of treat-
ment with savings generated by preventing unintended
pregnancy. All costs are given in Canadian dollars.
Ordinary oral contraceptives that contain estrogen
From the Office of Population Research, Princeton University, Prince-
ton, New Jersey, Department of Family Practice, University of British
Columbia, Vancouver, British Columbia, and Department of Obstetrics
and Gynecology, Laval University, Ste-Foy, Quebec, Canada.
VOL. 97, NO. 5, PART 1, MAY 2001
and progestin and pills containing progestin only can
be used for emergency contraception. Several brands of
combined oral contraceptives, one dedicated combina-
tion product (Preven; Shire Canada, Inc., Oakville,
Ontario), and one dedicated progestin-only product
(Plan B; Paladin Labs, Inc., Montreal) are available in
Canada. All require a prescription and are available in
family planning clinics and through general practitio-
ners and gynecologists. Use of combined pills reduces
the risk of pregnancy by about 75%,2– 4 whereas use of
progestin-only pills reduces the risk by 89%.5
Emergency contraception has the potential to reduce
significantly the incidence of unintended pregnancy
and the consequent need for abortion.6 In Canada, more
than 100,000 women each year have abortions.7
Methods
We analyzed cost savings from two different emergency
contraceptive protocols: (1) women seeking treatment
after unprotected intercourse, and (2) providing women
with emergency contraceptive pills during routine vis-
its to a clinician for later use if unprotected intercourse
occurs.6 We considered only costs associated with med-
ical care and ignored social, psychologic, and other
nonmedical costs. Whether emergency contraceptive
pills were cost-saving to the payer (the Ministry of
Health) depended on cost of treatment, probability of
preventing an unintended pregnancy, and cost of such
a pregnancy. Cost data were obtained from the British
Columbia Ministry of Health. Each province adminis-
ters its medical insurance plan slightly differently, but
we chose to use one province’s data for convenience.
The average cost of an unintended pregnancy is
$1289 (Table 1), which represents the weighted average
of the costs of ectopic pregnancy, induced abortion,
spontaneous abortion, and birth, with weights (1%,
51%, 12%, and 37%, respectively) equal to the probabil-
ities of each outcome for unintended pregnancies.8
0029-7844/01/$20.00 789
PII S0029-7844(01)01333-3
Table 1. Cost* of Unintended Pregnancy and Emergency
Contraceptive Pills
Average cost of an unintended pregnancy,
averted-birth model†
Average cost of an unintended pregnancy,
delayed-birth model‡
Average cost of a birth
Average cost of a spontaneous abortion
Average cost of an ectopic pregnancy
Average cost of an induced abortion
Cost of Preven
Cost of Plan B
Cost of physician visit
* Costs are in Canadian dollars.
†
Assumes that an unintended (either unwanted or mistimed) birth,
if avoided today, will never occur later.
‡
Assumes that an unwanted birth, if avoided today, will never
occur later but that a mistimed birth, if avoided today, will occur 2
years later.
Estimates in our base model, hereafter denoted the
averted-birth model, overstate the medical costs of
unintended pregnancies because we assumed that
births, if prevented today, would not occur later. In fact,
only 30% of unintended births, if prevented, will not
occur later; the remainder are timing failures (births
that occur sooner than intended).9 Therefore, we also
developed a delayed-birth model, assuming that
women with timing failures would become pregnant
intentionally 2 years later and discounting costs beyond
the current year to present value at an annual rate of
5%. In the delayed-birth model, the average cost of an
unintended birth decreased to $869 (⫽$2381 ⫺ [1.0 ⫺
0.70/1.052]), and the average cost of an unintended
pregnancy consequently decreased to $735. The average
cost of treating women after unprotected intercourse
was $31.53 for Preven ($26.53 for the office visit and
$5.00 for the drug) and $42.48 for Plan B ($26.53 for the
office visit and $15.95 for the drug).
The cost of advance provision depends directly on
the number of times women use emergency contracep-
tion. We assumed that women using barrier methods,
spermicides, periodic abstinence, and withdrawal rou-
tinely are provided three packs of either product.6 (We
ignored treatment among women using oral contracep-
tives who miss pills.10) The cost per woman receiving
Preven would be $41.53 ($26.53 for the office visit and
$15.00 for the drug), and the cost per woman receiving
Plan B would be $74.38 ($26.53 for the office visit and
$47.85 for the drug). For the diaphragm and the cervical
cap, costs would be lower ($15.00 for Preven and $47.85
for Plan B) because no extra office visit would be
required beyond the one to fit the device.
Women who rely on male or female condoms could
use an emergency regimen whenever condoms were
used imperfectly (not used or used incorrectly) or were
790 Trussell et al Emergency Contraception in Canada
used perfectly but nevertheless broke or slipped. In
contrast, women who rely on other methods would use
$1,288.61 emergency contraception only when they did not use
their contraceptive or used it incorrectly. Only at such
$734.79
times would they know that intercourse was unpro-
$2,381.39 tected. Therefore, we assumed that use of combined
$742.78 and progestin-only pills would prevent 74% and 89%,
$1,580.46 respectively, of pregnancies conceived during typical
$618.46 (perfect and imperfect) use of condoms and imperfect
$5.00
use of other methods (Table 2). For all methods except
$15.95
$26.53 the sponge, typical-use and perfect-use probabilities
were taken from a literature review.11 The typical-use
probability for the sponge was taken from results of a
clinical trial (Colin P, Creatsas G, Guerrero E, Guilbert
E, Guimont J, Serfaty D. Multinational open phase II
study of the efficacy, safety and acceptability of the
Protectaid contraceptive sponge. Presented at the 55th
Annual Conference of the Society of Obstetricians and
Gynecologists of Canada. June 1999, Ottawa, Ontario,
Canada). However, in this study the perfect-use prob-
ability (0.11) was estimated incorrectly and therefore
biased downward because all exposure was included in
the denominator, not just exposure during correct and
consistent use of the method. Therefore we arbitrarily
set the perfect-use probability at 0.13. Given that
women do not use their regular contraceptive perfectly,
they might not use emergency contraception after every
incidence of unprotected intercourse. Therefore, we
modeled two scenarios, in which emergency contracep-
tives are used after all and after three quarters of all
unprotected acts of intercourse.
Results
In the averted-birth model, one use of Preven saves
$2.35 in unintended pregnancy costs for every dollar
spent on the drug (Table 3), whereas Plan B saves $2.09
per dollar spent. In the delayed-birth model, a single
treatment with Preven or Plan B is still cost-saving. In
the averted-birth model, advance provision and consis-
tent use of Preven would produce annual savings from
$2.90 among women who use the sponge to $12.23
among parous women who use the cervical cap (Table
4) for each dollar spent on emergency contraception.
With less consistent use, savings would decrease by
25%, but savings for each dollar spent would be at least
$2.17. Advance provision of Plan B saves less because
the higher price of the drug outweighs its higher
efficacy. Nevertheless, advance provision of Plan B is
cost-saving.
In the delayed-birth model, annual savings per dollar
spent on Preven (provided in advance) would range
from $1.65 among women using the sponge to $6.97
among parous users of the cervical cap, when the
Obstetrics & Gynecology
Table 2. Annual Probabilities of Pregnancy During Typical and Perfect Use and With Emergency Contraceptive Backup
Emergency contraceptive backup*

Method use Consistent use† Inconsistent use‡

Method Typical Perfect Preven Plan B Preven Plan B

Male condom 0.14 0.03 0.04 0.02 0.06 0.05

Female condom 0.21 0.05 0.05 0.02 0.09 0.07
Diaphragm 0.20 0.06 0.09 0.07 0.12 0.10
Cervical cap
Parous women 0.40 0.26 0.26 0.23 0.29 0.27
Nulliparous women 0.20 0.09 0.11 0.09 0.13 0.11
Sponge 0.23 0.13 0.14 0.12 0.16 0.15
Spermicides 0.26 0.06 0.10 0.07 0.14 0.12
Withdrawal 0.19 0.04 0.07 0.05 0.10 0.08
Periodic abstinence 0.25 0.09 0.12 0.09 0.15 0.13
* We assumed that for methods other than condoms, pregnancy rates during imperfect use would be reduced by E% if emergency contraceptive
pills are used consistently (where E ⫽ 74% for combined pills and E ⫽ 89% for progestin-only pills); therefore, the pregnancy rate during typical
use of a method combined with consistent use of emergency contraception would be Ppfp ⫹ (1 ⫺ E) ⫺ (ft ⫺ Ppfp), where ft and fp are contraceptive
failure rates during typical and perfect use, respectively, and Pp is the proportion of use that is perfect, here assumed conservatively to be 80%.
Estimates of ft and fp are drawn from the literature on contraceptive efficacy.11 When emergency contraceptive pills are used after only
three-quarters of all unprotected acts of intercourse, the pregnancy rate during typical use of a method would be Ppfp ⫹ (1.0 ⫺ 0.75 ⫺ E)(ft ⫺ Ppfp).
We ignored the distinction between annual probabilities of failure (ft and fp) and annual failure rates because the two are nearly identical. Consider
the following simple birth-interval model for women using spermicides. An annual probability of failure during typical use of 0.21 implies an
average monthly probability of failure of 0.0195 (⫽ 1.0 ⫺ [1.0 ⫺ 0.21]1/12). Assuming this probability is constant over time, the average waiting time
to conception while using spermicides is 51 months (⫽ 1/0.0195). Then ignoring ectopic pregnancies, spontaneous abortions, and stillbirths and
assuming that every other pregnancy ends in induced abortion, the waiting time to a conception leading to a live birth would be 106 months: 51
months to get pregnant the first time, three months’ gestation until the abortion, 1 month of postpartum nonsusceptibility after the abortion, and
51 months to get pregnant again. The entire interval from one birth to the next would be 117 months: 2 months of postpartum nonsusceptibility
after the birth (assuming minimal breastfeeding), 106 months of waiting time to the next live-birth conception, and 9 months for gestation. Hence
a birth occurs every 117 months or 9.75 years, so the birth rate per year is 0.103 (⫽ 1/9.75). Because there are two pregnancies for each birth, the
pregnancy rate per year is 0.206 (⫽ 2 ⫺ 0.103), a rate that is close to the annual probability of failure of 0.21. Differences are even smaller for more
effective methods.
†
Assumes use 100% of the time when needed.
‡
Assumes use only 75% of the time when needed.

regimen is used consistently. Less consistent use would Discussion

still be cost-saving among users of all methods. Ad-
vance provision of Plan B is cost-saving whether the Emergency contraception reduces expenditures on
drug is used consistently or inconsistently, except medical care by preventing unintended pregnancies,
among women who rely on the male condom or the which are expensive. It is generally cost-saving whether
sponge. it is provided when the emergency occurs or in advance
to be used as needed. Several caveats are in order when
interpreting our results. We used data from the United
States on contraceptive failure, outcomes of unintended
Table 3. Costs* and Cost Savings for Single Use of pregnancies, and proportion of unintended births that
Emergency Contraceptive Pills
are unwanted because Canadian data were not avail-
Averted-birth Delayed-birth able. The proportion of unwanted among unintended
model† model‡
births might be lower in Canada because unwanted
Savings/ Savings/ births are a larger fraction of unintended births among
ECP regimen Cost Savings cost Savings cost
the poor and blacks,9 and Canada has proportionately
Preven $31.53 $74.10 $2.35 $42.25 $1.34 fewer women in these groups than does the United
Plan B $42.48 $88.79 $2.09 $50.63 $1.19
States. Contraceptive failure rates during typical use
ECP ⫽ emergency contraceptive pills.
might be lower in Canada for the same reason.12,13
* Costs are in Canadian dollars.
†
Assuming that an unintended (either unwanted or mistimed) birth, Moreover, our estimates of cost savings from emer-
if avoided today, will never occur later. gency contraceptive pills are likely to be understated for
‡
Assuming that an unwanted birth, if avoided today, will never
occur later but that a mistimed birth, if avoided today, will occur 2 six reasons. First, in the scenario in which emergency
years later. contraception is provided only after unprotected inter-

VOL. 97, NO. 5, PART 1, MAY 2001 Trussell et al Emergency Contraception in Canada 791
Table 4. Costs* and Cost Savings for Advance Provision of Emergency Contraceptive Pills
Averted-birth model†—Preven Delayed-birth model‡—Preven

Inconsistent use of Inconsistent use of

Consistent use of ECPs§ ECPs¶ Consistent use of ECPs§ ECPs¶

Method Cost Savings Savings/cost Savings Savings/cost Savings Savings/cost Savings Savings/cost

Male condom $41.53 $133.75 $3.22 $100.31 $2.42 $76.27 $1.84 $57.20 $1.38
Female condom $41.53 $200.63 $4.83 $150.47 $3.62 $114.40 $2.75 $85.80 $2.07
Diaphragm $15.00 $145.22 $9.68 $108.91 $7.26 $82.81 $5.52 $62.10 $4.14
Cervical cap
Parous $15.00 $183.43 $12.23 $137.57 $9.17 $104.60 $6.97 $78.45 $5.23
Nulliparous $15.00 $122.29 $8.15 $91.72 $6.11 $69.73 $4.65 $52.30 $3.49
Sponge $41.53 $120.38 $2.90 $90.28 $2.17 $68.64 $1.65 $51.48 $1.24
Spermicides $41.53 $202.54 $4.88 $151.90 $3.66 $115.49 $2.78 $86.62 $2.09
Withdrawal $41.53 $150.95 $3.63 $113.21 $2.73 $86.07 $2.07 $64.56 $1.55
Periodic abstinence $41.53 $170.06 $4.09 $127.54 $3.07 $96.97 $2.33 $72.73 $1.75

Averted-birth model†—Plan B Delayed-birth model‡—Plan B

Male condom $74.38 $160.20 $2.15 $120.15 $1.62 $91.35 $1.23 $68.51 $0.92
Female condom $74.38 $240.30 $3.23 $180.23 $2.42 $137.02 $1.84 $102.77 $1.38
Diaphragm $47.85 $173.93 $3.63 $130.45 $2.73 $99.18 $2.07 $74.38 $1.55
Cervical cap
Parous $47.85 $219.70 $4.59 $164.78 $3.44 $125.28 $2.62 $93.96 $1.96
Nulliparous $47.85 $146.47 $3.06 $109.85 $2.30 $83.52 $1.75 $62.64 $1.31
Sponge $74.38 $144.18 $1.94 $108.14 $1.45 $82.21 $1.11 $61.55 $0.83
Spermicides $74.38 $242.59 $3.26 $181.94 $2.45 $138.33 $1.86 $103.75 $1.39
Withdrawal $74.38 $180.80 $2.43 $135.60 $1.82 $103.09 $1.39 $77.32 $1.04
Periodic abstinence $74.38 $203.68 $2.74 $152.76 $2.05 $116.14 $1.56 $87.11 $1.17
ECPs ⫽ emergency contraceptive pills.
* Costs are in Canadian dollars.
†
Assuming that an unintended (either unwanted or mistimed) birth, if avoided today, will never occur later.
‡
Assuming that an unwanted birth, if avoided today, will never occur later but that a mistimed birth, if avoided today, will occur 2 years later.
§
Assuming ECPs are used 100% of the time when needed.
¶
Assuming ECPs are used only 75% of the time when needed.

course, we assumed that an office visit is always re-
quired. In fact, some clinicians prescribe by telephone,
especially for established patients, or delegate to a
nurse because a physical examination is not medically
necessary. When an emergency regimen is prescribed in
advance, women can renew their prescription by tele-
phone as needed.
Second, we assumed that women or providers would
pay average wholesale prices for pills. In fact, many
providers receive discounts. For example, some pay
only $3.25 for an entire pack of Ovral (Wyeth-Ayerst
Canada, Inc., St-Laurent, Quebec) (enough for five
treatments), whereas we assumed they would provide
Preven, which costs $5.00 for a single treatment.
Third, we assumed that treatment after unprotected
intercourse prevented 74% (combined pills) or 89%
(progestin-only pills) of pregnancies that would oth-
erwise occur. The true effectiveness of these regimens
is probably higher because the observed number of
pregnancies is likely to be too high and the expected
number of pregnancies is probably too low in efficacy
studies.2– 4
Fourth, when we examined advance provision, we
assumed in the worst-case scenario that emergency
contraceptive pills would be used on only three quar-
ters of occasions when they were needed. Women and
couples who consistently do not use a regular method
of contraception might use emergency regimens far
more consistently. One might argue instead that a
woman who knows she has a backup will become less
diligent with her regular method; however, if used as
ongoing methods, emergency regimens would be far
less effective than most other contraceptives. During a
year of typical use, a woman’s risk of pregnancy would
be 20% with Plan B and would exceed 35% with Preven.
One in two women have nausea and one in five women
vomit after using the combination regimen, and using it
repeatedly in one cycle or in several cycles in a row
would result in menstrual chaos. Those risks are likely
to dissuade women from frequent use and might en-
hance regular contraceptive use. Making emergency
contraception more widely available does not increase
risk-taking but reduces the incidence of unintended
pregnancy.14 Women who are the most diligent about

792 Trussell et al Emergency Contraception in Canada Obstetrics & Gynecology

ongoing contraceptive use are those most likely to seek tive pills: A simple proposal to reduce unintended pregnancies.
emergency treatment.15 Fam Plann Perspect 1992;24:269 –73.
7. Statistics Canada. Therapeutic abortions, 1995. Health Reports
Fifth, when we examined advance provision, we 1997;9(3):55.
assumed that each woman would use her emergency 8. Henshaw SK. Unintended pregnancy in the United States. Fam
regimen three times. Use of emergency contraception Plann Perspect 1998;30:24 –29, 46.
three or more times in a year is very uncommon,14 so 9. Abma JC, Chandra A, Mosher WD, Peterson LS, Piccinino LJ.
the average annual cost of advance provision is over- Fertility, family planning, and women’s health: New data from the
1995 National Survey of Family Growth. Vital Health Stat 1997;23:
stated. If each woman used only two packs of emer- 25.
gency contraceptive pills on average, then advance 10. Guillebaud J. When do you need to give emergency contraception
provision of Plan B would be cost-saving even under for missed pills? Fertil Control Rev 1995;4:18 –20.
the least favorable assumptions that mistimed births are 11. Trussell J. Contraceptive efficacy. In: Hatcher RA, Trussell J,
merely delayed for 2 years and that the regimen is used Stewart F, Cates W, Stewart GK, Kowal D, Guest F, eds. Contra-
ceptive technology. 17th rev ed. New York: Ardent Media, 1998;
only 75% of the time when needed. 779 – 844.
Sixth, we underestimated cost of births resulting from 12. Trussell J, Vaughan B. Contraceptive failure, method-related dis-
unintended pregnancy. We assumed that unintended continuation and resumption of use: Results from the 1995 Na-
births that were not unwanted would be postponed for tional Survey of Family Growth. Fam Plann Perspect 1999;31:64 –
2 years and would be intended at that time. The average 72, 93.
13. Fu H, Darroch JE, Haas T, Ranjit N. Contraceptive failure rates:
period that such births would be postponed is likely to New estimates from the 1995 National Survey of Family Growth.
be much longer; and the longer such births are post- Fam Plann Perspect 1999;31:56 – 63.
poned, the more likely it is that they will never occur. 14. Glasier A, Baird D. The effects of self-administering emergency
We incorrectly assumed that the cost of medical care for contraception. N Engl J Med 1998;339:1– 4.
an unintended birth is no greater than the average cost 15. Kosunen E, Sihvo S, Hemminki E. Knowledge and use of hormonal
emergency contraception in Finland. Contraception 1997;55:153–7.
for all births.16 16. Brown SS, Eisenberg L, eds. The best intentions: Unintended
pregnancy and the well-being of children and families. Washing-
ton DC: National Academy Press, 1995.
References
1. Trussell J, Koenig J, Ellertson C, Stewart F. Preventing unintended
pregnancy: The cost-effectiveness of three methods of emergency Address reprint requests to:
contraception. Am J Public Health 1997;87:932–7. James Trussell, PhD
2. Trussell J, Ellertson C, Stewart F. The effectiveness of the Yuzpe Office of Population Research
regimen of emergency contraception. Fam Plann Perspect 1996;28: Wallace Hall
58 – 64, 87. Princeton University
3. Trussell J, Rodriguez G, Ellertson C. New estimates of the effec- Princeton, NJ 08544
tiveness of the Yuzpe regimen of emergency contraception. Con-
E-mail: trussell@princeton.edu
traception 1998;57:363–9.
4. Trussell J, Rodriguez G, Ellertson C. Updated estimates of the
effectiveness of the Yuzpe regimen of emergency contraception.
Contraception 1999;59:147–51. Received July 31, 2000.
5. Task Force on Postovulatory Methods of Fertility Regulation. Received in revised form October 27, 2000.
Randomised controlled trial of levonorgestrel versus the Yuzpe Accepted January 31, 2001.
regimen of combined oral contraceptives for emergency contracep-
tion. Lancet 1998;352:428 –33. Copyright © 2001 by The American College of Obstetricians and
6. Trussell J, Stewart F, Guest F, Hatcher RA. Emergency contracep- Gynecologists. Published by Elsevier Science Inc.

VOL. 97, NO. 5, PART 1, MAY 2001 Trussell et al Emergency Contraception in Canada 793