Health-related quality of life changes among users of depot medroxyprogesterone acetate for contraception
Health-related quality of life changes among users of depot
medroxyprogesterone acetate for contraception☆,☆☆
Sikolia Z. Wanyonyi⁎, William R. Stones, Evan Sequeira
Department of Obstetrics and Gynaecology, Aga Khan University Hospital, Nairobi; 3rd Parklands Avenue, Box 30270, Nairobi 00100, Kenya
Received 27 March 2011; revised 27 May 2011; accepted 28 May 2011
Background: Depot medroxyprogesterone acetate (DMPA) may have other noncontraceptive effects that could impact on the quality of life. The objective of this study was to assess the health-related quality of life changes associated with the use of DMPA for contraception. Study Design: A prospective, observational study using the Short Form-36 quality of life questionnaire. Results: After 6 months of use, the participants had an improved physical summary score, mean change [5.64 (95% confidence interval [CI],1.87–9.4), p=.054]. There was no significant change in sexual function [5.33 (95% CI, −2.15 to 12.81), p=.0858] and mental summary score[−0.51 (95% CI, −1.90 to 2.92), p=.432]. The main side effect of DMPA was menstrual irregularity (32.5%); 17.2% of the participants foundamenorrhea desirable. Conclusion: Besides its contraceptive efficacy, DMPA is associated with an improvement in perceived physical health with no apparentadverse effect on mental health and sexual function. 2011 Elsevier Inc. All rights reserved.
Keywords: Depo medroxyprogesterone acetate; Health-related quality of life; Contraception; SF-36
Similar to other hormonal contraceptives, users of DMPA
may experience various side effects. Most of these side
Depot medroxyprogesterone acetate (DMPA) is a medi-
effects are related to the suppression of ovarian estradiol
um-term reversible contraceptive method, with a low failure
production. These include unscheduled bleeding and amen-
rate, 0.3 to 3 per 100 woman-years . However, its effec-
orrhea. Minor side effects include nausea, dizziness, weight
tiveness for child spacing in sub-Saharan Africa has been
gain, hair loss, acne and headaches. Accompanying mood
questioned, though its popularity remains high in the region
changes related to some of these side effects are documented
. There have been reports linking the use of DMPA with
. Alteration in the sexual drive among DMPA users has
an increased risk of HIV transmission, but recent work has
been reported. These effects are, however, reversible upon
reliably disapproved possible association Besides con-
discontinuation of the method . There have also been
traceptive effects, DMPA also possesses other benefits, for
concerns about the effects of DMPA on the bone mineral
example, reduction in risk of endometrial carcinoma, reduced
density. This effect is minimal and reversible but may be
incidence of iron deficiency anemia and dysmenorrheal . Its
associated with early onset osteoporosis if DMPA is used
action on the cervical mucus has also been thought to be
protective against pelvic inflammatory disease
All these side effects have varying influence on the
acceptability of this method for contraception. Irregularbleeding, for instance, accounts for most discontinuations inthe first 6–9 months of use These side effects and
their influence on discontinuation rates has led to focus on
☆☆ Funding: We received funding from the Aga Khan University
contraceptive research being mainly on the burden of gyne-
research grant for postgraduate medical education.
cological and other related medical conditions, with little
⁎ Corresponding author. P.O. Box 16963-00100 Nairobi, Kenya
attention on how they affect the user's quality of life (QoL).
It is now widely acknowledged that the personal burden of
0010-7824/$ – see front matter 2011 Elsevier Inc. All rights reserved.
S.Z. Wanyonyi et al. / Contraception xx (2011) xxx–xxx
illness cannot be described fully by measures of disease
At 6-month follow-up, the questionnaire was given to the
status alone. Psychosocial factors such as apprehension,
respondents prior to the clinical consultation. Participants
functional impairment, difficulty in fulfilling personal and
were also reminded by a telephone call on their scheduled
family responsibilities, financial burden and diminished
appointment in the event that they did not return as expected.
cognition must also be encompassed . Some lifestyle
The SF-36 was completed in the same manner as before,
choices such as contraception may have effects on daily life
and information on experience with DMPA and any side
and life satisfaction besides providing fertility control. These
effects, either desirable or unwanted, was sought during the
effects could be evaluated using either the existing generic
consultation. In case of discontinuation, the reasons were
instruments or condition-specific QoL assessment tools.
There are several generic health-related QoL tools in exis-
tence, and these have been widely used in other disciplineswith reliable outcomes .
Consecutive sampling technique was used to recruit study
This study aims to assess the health-related QoL changes
participants. Every participant meeting the inclusion criteria
among Kenyan women using DMPA for contraception by
was approached for possible recruitment to the study after
utilizing the Short Form (SF)-36 questionnaires besides
evaluating for other side effects associated with the method.
shows the flow of the study. Estimates of sample size were derived from the SF-36
health survey manual by Ware . These are predetermined
sample size calculations based on a healthy US populationand widely used in QoL studies utilizing the SF-36. A
A prospective, observational study was carried out at the
sample size of 105 participants was needed to detect a
Aga Khan University Hospital and the Family Health
5-point difference in the mental summary score (MCS)
changes over time within the same group. A power calcu-
The study was conducted for a period of 10 months
lation of 80% was used given a two-tailed t test with an
starting from December 2008 to September 2009.
α value of .05 and an intertemporal correlation between
The broad objective of our study was to assess the health-
scores of 0.60. A 5-point difference was deemed to be both
related QoL changes among women using DMPA for
contraception. The specific objectives are as follows: (i) to
We included women who were aged 18–49 years and
determine the characteristics of women using DMPA for
willing to give informed consent and to continue follow-up
contraceptive, (ii) to establish the reasons for choosing to use
at our study clinics. All participants had to fulfill the World
or discontinue the method, (iii) to evaluate the changes in
Health Organization medical eligibility criteria for the
QoL among women using DMPA and (iv) to explore the side
Any woman who had used DMPA within the previous
12 months at the time of recruitment or did not meet theWorld Health Organization eligibility criteria for use of
All clients desiring contraceptives were counseled on the
DMPA was excluded from the study. Those with history of
available methods including DMPA. This was done in both
chronic illness, current menstrual abnormalities or previous
clinics by a family planning provider. The method chosen by
mental illness were also excluded. Women who were less
the client was then administered in the usual way.
than 6 months postpartum were not eligible for the study.
Those who chose DMPA were approached for recruit-
ment into the study. If they met the eligibility criteria and
agreed to participate, a brief description of the study was readto them and an informed consent obtained. The socio-
The primary outcome measure was the MCS of SF-36.
demographic data were captured, and a brief clinical history
Other outcome measures included physical summary score
was sought. The SF-36 was then introduced to the client for
(PCS) and sexual function summary score. The secondary
self-administration. The questionnaires were completed in
outcome measures were discontinuation rates for DMPA,
the clinic. Any woman requesting to complete the question-
reasons for initiation and discontinuation of DMPA and side
naire later was requested to return it at the earliest
opportunity. An addressed, stamped envelope was provided
when desired. However, only four participants opted for thisalternative, and three of them failed to return the question-
Data were collected using a self-administered 36-item
naires despite constant reminders. They were consequently
SF-36 health survey at 0 and 6 months, questionnaire for
sociodemographic variables and clinical history and a ques-
Upon returning the questionnaire, it was checked for
tionnaire for follow-up on 6-month experience with DMPA.
completeness. Difficulties encountered during completion
Sexual function data were collected using four items
were ascertained. Assistance was provided as desired.
derived from the medical outcomes survey questionnaire
S.Z. Wanyonyi et al. / Contraception xx (2011) xxx–xxx
147 women screened 16 declined 131 recruited 24 excluded (not eligible) 107 met eligibility criteria 9 Lost to follow-up Follow up 98 included in the final 6-month analysis
Quantitative data were entered into the statistical software
intervals (CI) for normally distributed data. The median
spreadsheet. Incomplete SF-36 forms were excluded from
value with interquartile range (IQR) was used to describe
nonparametric data. A comparison of the medians was per-formed using the Wilcoxon rank test and the Student's t test
for means. Subanalysis was done using multivariate
Consecutive sampling was used, giving all the users an
regression models for probable confounders.
equal chance to participate in the study. This minimized selec-tion bias and intentional recruitment of study participants.
On follow-up, the respondents were required to com-
The study was approved by the Aga Khan University
plete the questionnaires before being attended to by a family
ethics and research committee. Informed written consent was
planning provider. This was to ensure that the response they
gave was not influenced in any way by the consultation.
Confidentiality and anonymity were maintained through-
2.6. Scoring and calculation for the SF-36
out the study by identifying participants by their uniquemedical record numbers only.
Scoring for the SF-36 was performed using the RAND-
36-item health survey technique A three-step approachwas used involving (i) standardization of scale, (ii)
aggregation of scale scores and (iii) transformation ofsummary scores in line with recommendation from the SF-
A total of 147 participants were screened, of which 131
36 manual. This method described by Ware is
were recruited to the study, but only 107 met the eligibility
reproducible with a reliability coefficient of 0.78–0.93.
criteria and were assigned to the 6-month follow-up. Of
The advantage of using the PCS and MCS is that it reduces
these, 98 (91.6%) completed the study (see ).
the SF-36 to two summary measures without substantial lossof information.
The participants had a mean age of 30.7 (SD, 5.5) years.
Analysis was performed using the SPSS® version 15 and
The average family size was 1.7 (SD, 1.1), with the desired
STATA® version 10. Statistical tests were two-sided, and a
number of children being 2.7 (SD, 1.1). The mean age of
statistical test with a p value of b.05 was considered
first pregnancy was 22.4 (SD, 8.4) years; 10.3% of the
statistically significant. Comparison between means was
respondents were nulliparous. The other sociodemographic
undertaken using the Student's t test with 95% confidence
S.Z. Wanyonyi et al. / Contraception xx (2011) xxx–xxx
The majority of the participants (72%; 95% CI, 47.2–
62.2) were contraceptively naive. presents the dif-
ferent methods the clients had used 12 months prior to
The reasons for opting to use DMPA varied, with 72%
(95% CI, 65.3–80.6) choosing it for convenience ().
a IUD, intrauterine contraceptive device. b Others included Chinese pills, traditional methods and herbs.
The PCS improved significantly over the 6-month period,
with no changes noted in the MCS. There was an
3.6. Reasons for discontinuing the method
improvement in the sexual function score, though this wasnot statistically significant (
A total of 15 participants (15.3%) discontinued DMPA
The various changes in the individual HRQL component
within the study period (4 participants at 3 months and 11 at
scores are presented in . There was improvement in
6 months of follow-up). The main reasons for discontinuing
the physical functioning [baseline vs. 6-month median score
the method were menstrual irregularity (26.6%), reduced
(IQR), 90 (20) vs. 95 (15); p=.0235; general health, 75 (25)
libido (13.3%), need for longer acting method (20%) and
vs. 85 (15); p=.0011]. The other scales did not show any
weight gain (20%). Only 6.7% of the participants discon-
tinued the method following medical advice.
There was no significant influence of sociodemographic
characteristics on the summary scores. Primary level of
3.7. Participants’ satisfaction with DMPA
education had a small influence to the change in PCS, but
shows the reasons for satisfaction with the use
there were only three respondents (2.8%; 95% CI, 0.0-6.0)
with a primary level of education; hence, this effect did notsignificantly affect the results. There was also no influence ofmarital status, presence of dysmenorrhea and prior use of
contraceptives on both the PCS and MCS changes.
In this prospective study, women using DMPA for
contraception had improved physical functioning and
After 6 months of use of DMPA, 40 participants (40.8%;
general health after only 6 months. There was a general
95% CI, 38.6-49.2) reported undesirable side effects, while
improvement in the PCS among users of DMPA, with the
58 respondents (59.2%; 95% CI, 52.6-67.3) did not
primary outcome of measure (MCS) remaining the same.
This confirmed our null hypothesis for the primary outcomemeasure. There was also an improvement in sexual function,though this change was not statistically significant. Howev-
er, we found this to be a clinically relevant finding in our
study as DMPA has been associated with a reduction in
sexual arousal, libido and pleasure . Based on the
CIs around the mean changes seen, the possible effects of
Tertiary includes any education beyond secondary but excluding
S.Z. Wanyonyi et al. / Contraception xx (2011) xxx–xxx
Six-month mean changes in the summary scale score
50.9 (16.6) 56.5 (23.7) 5.64 (1.87 to 9.40)
50.6 (13.5) 50.1 (15.8) −0.51 (−1.90 to 2.92)
76.0 (37.5) 81.3 (43.7) 5.33 (−2.15 to 12.81) .0858
DMPA could have ranged from 2-point deterioration in the
sexual function scale to a 13-point improvement (95% CI,−2.15 to 12.81). We could therefore assert based on ourresults that DMPA did not adversely affect sexual function
reports of adverse events being higher. Important to note is
that the 12-month discontinuation rate for injectable
The improvement in health-related QoL for women using
contraceptives reported in the Kenya demographic health
DMPA was encouraging for a method widely used in Kenya.
surveys was higher (29%); however, it was the family
Having recorded an improvement in the contraceptive
planning method with the least discontinuation rate . It
prevalence rate from 39% to 46% among women aged 15–
therefore follows that the presence of unwanted side effects
49 years between 2004 and 2009 and with an upward trend in
does not necessarily equate to discontinuation regardless of
uptake of injectable contraceptives , our results, though
the population being studied. Most women chose the
not generalizable, should be encouraging for users of
method for convenience and found it so after 6 months of
DMPA. This reflects the fact that besides its contraceptive
use. The added advantage of improvement in certain
efficacy, DMPA could possess other noncontraceptive
aspects of QoL could also have enhanced its continued use
benefits that improve the QoL. These outcomes were not
influenced by sociodemographic characteristics like age,
This study had several limitations. It was restricted to an
level of education, occupation, marital status and previous
urban population with a high level of education and pre-
sumably a higher socioeconomic status in a country where
Women can realize their reproductive goals only when
more than 80% of the reproductive health population is rural
they use contraceptive methods continuously. A prominent
. It may be difficult to extrapolate the results to other users
concern for most family planning providers is the
discontinuation of methods. In our study, despite 40.8%
However, this study was adequately powered to detect a
of the women reporting adverse side effects with use of
5-point difference that was considered clinically relevant
DMPA, this was not reflected in the discontinuation rate
with an 80% precision. It was also prospective in design with
(15.3%). The 6-month discontinuation rate is comparable
very minimal loss to follow-up. Up to 91.6% of the parti-
to that quoted in other studies Aruasa and
cipants completed the 6-month follow-up. The method of
Wanyoike found a 6-month discontinuation rate of
contraceptive chosen was the most preferred with the least
16% locally among users of DMPA despite subjective
discontinuation rate, hence reflecting a larger proportionof contraceptive users. Restriction to one method ensureduniformity of the outcomes since different contraceptive
methods have different effects. Despite being mainly urban,
A comparison of baseline and 6-month HQRLa component scores
the Kenya demographic and health surveys 2009 showed
that the majority (53%) of the contraceptive users were
HQRL, health-related quality of life.
S.Z. Wanyonyi et al. / Contraception xx (2011) xxx–xxx
married women in urban areas. Furthermore, modern
methods use was generally higher in urban (47%) than inrural areas, and more than 60% of the contraceptive users had
 Fu H, Darroch JE, Haas T, Ranjit N. Contraceptive failure rates: new
at least some secondary level of education The results
estimates from the 1995 National Survey of Family Growth. FamPlann Perspect 1999;31:56–63.
of this study could therefore still be descriptive of the
 MEASURE DHS+. Demographic and Health Surveys. Quality informa-
pattern of modern Kenyan contraceptive users.
tion to plan, monitor and improve population, health, and nutrition pro-
We used generic tools that were tested and found to be
reliable for the study population, making the results com-
 Ngianga-Bakwin K, Stones RW. Birth intervals and injectable
parable to other QoL study results. However, caution needs
contraception in sub-Saharan Africa. Contraception 2005;71:353–6.
 Lavreys L, Baeten JM, Martin H, et al. Hormonal contraception and
to be exercised in interpreting our results as the SF-36 may
risk of HIV-1 acquisition: results of a 10-year prospective study. AIDS
sometime produce different results from those of primary
 Morrison CS, Richardson BA, Mmiro F, et al. Hormonal contraception
However, these results are clinically relevant for fam-
and the risk of HIV acquisition (HC-HIV) study group. AIDS 2007;21:
ily planning providers faced with the task of counseling
 Glasier A. Contraception. In: DeGroot LJ, & Jameson JL, editors.
women on noncontraceptives benefits of DMPA. Women
Endocrinology, 5th ed. Philadelphia: Elsevier Saunders; 2006.
will often be apprehensive on how the methods chosen will
impact on their general well-being and sexuality. Women
 Loose DS, Stancel GM. Estrogens and progestins. In: Brunton LL,
will need to balance between the contraceptive benefits of
Lazo JS, & Parker KL, editors. Goodman & Gilman's the pharma-
birth spacing and fertility control and other noncontraceptive
cological basis of therapeutics, 11th ed. New York: McGraw-Hill;2006. p. 1541–71.
benefits. Besides the unwanted side effects, women should
 Rivera R, Yacobson I, Grimes D. The mechanism of action of hor-
be informed about the possibility of improvement in physical
monal contraceptives and intrauterine contraceptive devices. Am J
Obstet Gynecol 1999;181(5 Pt 1):1263–9.
While there was no statistically significant change in
 Trussell J. Contraceptive efficacy. In: Hatcher RA, Trussell J, &
sexual function, the CIs tended more toward improved
Stewart FH, et al, editors. Contraceptive technology, 18th rev. ed. NewYork: Arden; 2004. p. 773–845.
sexual function as opposed to deterioration. This is relevant
 Hatcher RA. DMPA injections, implants, and progestin-only pills
in counseling women who are concerned about the effect of
(minipills). In: Hatcher RA, Trussell J, & Stewart FH, et al, editors.
DMPA on sexual function. The mental well-being of women
Contraceptive technology, 18th rev. ed. New York: Ardent Media; 2004.
on DMPA was also not affected over time, and this could be
encouraging for users fearful of the impact of the method on
 Santen RJ. Endocrinology of breast and endometrial cancer. In: Strauss
JF, & Barbieri RL, editors. Yen and Jaffe's reproductive endocrinol-
their psychological well-being. The presence of unwanted
ogy, 5th ed. Philadelphia: Elsevier Saunders; 2004. p. 787–809.
side effects should not deter one from choosing a method as
 World Health Organization. Hormonal contraception and bone health.
they could still derive other benefits from it.
The public health implication of our findings is that
DMPA might improve some aspects of health-related QoL in
 Trussell J, Vaughan B. Contraceptive failure, method-related discon-
tinuation and resumption of use: results from the 1995 National Survey
women. These gains can be factored in to a woman's choice
of Family Growth. Fam Plann Perspect 1999;31:64–72.
of contraception. This benefit must be weighed against the
 WHO. Progress in reproductive health research.
overall short- and long-term risks, which must be individ-
ualized based on medical and reproductive history. Further
 Muldoon MF, Barger SD, Flory JD, Manuck SB. What are quality of
research with either generic or method-specific tools is
life measurements measuring? BMJ 1998;316:542–55.
 Garrat AM, Ruta DA, Abdalla MI, Buckingham JK, Russell IT. The
needed to assess the impact of other contraceptives on QoL.
SF36 health survey questionnaire: an outcome measure suitable for
The studies should also address the wider rural and low
routine use in the NHS? BMJ 1993;306:1140–4.
socioeconomic populations. In light of the findings of this
 Garrat A, Schmidt L, Mackintosh A, Fitzpatrick R. Quality of life
study, more QoL consideration should be given to contra-
measurement: bibliographic study of patient assessed health outcome
 Stewart AL, Hays RD, Ware JE. The MOS short-form general health
survey: reliability and validity in a critical patients’ population. Med
 Ware JE. SF-36 health survey; manual and interpretation guide. Boston,
Massachusetts: The Health Institute, New England Medical Center, 1993.
This longitudinal study showed a significant improve-
 Medical outcomes study: measures of quality of life core survey.
ment in certain aspects of QoL associated with a method
widely used in this country. It shows that besides conferring
 Aruasa W, Wanyoike G. The effect of DMPA contraceptive on body
highly effective fertility control, contraceptives could also
weight and blood pressure among indigenous Kenyan women in KNH.
confer other unintended health benefits to its users.
Kenya Obstetric and Gynecological Society Conference proceedings,Nairobi, Kenya; 2006.
Contraceptive users should be informed of these additional
 Contopoulous-loannidis DG, Karvouni A, A-loannidis JP. Reporting
benefits with increased promotion of the noncontraceptive
and interpretation of SF-36 outcomes in randomized trial: systematic
MATERIAL SAFETY DATA SHEET The batteries are exempt articles and are not subject to the OSHA Hazard Communication Standard Requirement. This sheet is provided as technical information only. The information and recommendations set forth are made in good faith and are believed to be accurate as of the date of preparation. However, Contour Energy Systems makes no warranty expressed or Implie