Secondary Logo

Journal Logo

Side Effects and Health Benefits of Depot Medroxyprogesterone Acetate

A Systematic Review

Dianat, Shokoufeh DO; Fox, Edith MPH; Ahrens, Katherine A. PhD; Upadhyay, Ushma D. PhD, MPH; Zlidar, Vera M. MHS; Gallo, Maria F. PhD; Stidd, Reva L. MS, MBA; Moskosky, Susan MS; Dehlendorf, Christine MD, MAS

doi: 10.1097/AOG.0000000000003089
Contents: Contraception: Review
Free
SDC
ABOG MOC II

OBJECTIVE: Counseling about potential side effects and health benefits of contraceptive methods could facilitate continued method use and method satisfaction, yet no evidence-based compilation of side effects and benefits exists to aid such counseling. Among contraceptive methods in the United States, depot medroxyprogesterone acetate (DMPA) injectables have the highest discontinuation rates, and most discontinuation is attributable to side effects. This review examines the side effects and health benefits of DMPA to inform counseling.

DATA SOURCES: We searched PubMed, POPLINE, EMBASE, Web of Science, Campbell Collaboration Library of Systematic Reviews, the Cochrane Database of Systematic Reviews, the Cochrane Center Register of Controlled Trials, and ClinicalTrials.gov.

METHODS OF STUDY SELECTION: We included English-language studies published from 1985 to 2016 that enrolled healthy, nonbreastfeeding females aged 13–49 years at risk of unintended pregnancy, compared intramuscular or subcutaneous progestin-only injectables to a contemporaneous comparison group, and addressed at least one key question: 1) What side effects are associated with progestin-only injectable contraceptive use? 2) What health benefits are associated with progestin-only injectable contraceptive use? Study quality was assessed using criteria from the U.S. Preventive Services Task Force.

TABULATION, INTEGRATION, AND RESULTS: Twenty-four studies met inclusion criteria. None were randomized controlled trials. There were 13 prospective cohort, five retrospective cohort, four case–control, and two cross-sectional studies. Studies of moderate or high risk of bias suggest an association between DMPA use and weight gain, increased body fat mass, irregular bleeding, and amenorrhea. Inconsistent evidence exists for an association between DMPA use and mood or libido changes. Limited evidence exists for an association between DMPA use and decreased risk of cancers and tubal infertility.

CONCLUSION: Higher-quality research is needed to clarify DMPA's side effects and benefits. In absence of such evidence, patient-centered counseling should incorporate the available evidence while acknowledging its limitations and recognizing the value of women's lived experiences.

Evidence suggests an association between depot medroxyprogesterone acetate use and weight gain, irregular bleeding, and amenorrhea; evidence is inconsistent regarding changes in mood or libido.

University of California, San Francisco, San Francisco, California; the Office of Population Affairs, Department of Health and Human Services, Rockville, Maryland; Social Solutions International's USAID Global Health Support, Initiative-III, Arlington, Virginia; The Ohio State University, College of Public Health, Division of Epidemiology, Columbus, Ohio; and Atlas Research LLC, Washington, DC.

Corresponding author: Shokoufeh Dianat, DO, University of California, San Francisco, 1001 Potrero Ave, Ward 6D, San Francisco, CA 94110; email: shokoufeh.dianat@gmail.com.

This review was funded, in part, by the following contracts between the Office of Population Affairs and Atlas Research Inc.: HHSP233201500126I, HHSP233201450040A.

Financial Disclosure The authors did not report any potential conflicts of interest.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Office of Population Affairs, Office of the Assistant Secretary for Health, or Department of Health and Human Services.

Each author has confirmed compliance with the journal's requirements for authorship.

Peer reviews are available at http://links.lww.com/AOG/B254.

Despite the availability of a wide range of contraceptive methods, many sexually active women in the United States who do not desire pregnancy do not use contraception. Most contraceptive discontinuation is attributed to perceived method-related side effects.1 In addition, most women cite side effects as “extremely important” factors in their contraceptive decisions.2,3

Anticipatory counseling about side effects has been associated with greater method satisfaction4 and continuation.5,6 Guidelines support providing personalized information to enable women to select the most appropriate method for them, based on health conditions or preferences about side effects.7

Unfortunately, women are frequently dissatisfied with information received during contraceptive counseling,8–10 possibly because of a lack of easily accessible, accurate information on potential side effects. No evidence-based, synthesized body of work exists in the United States regarding contraceptive side effects and health benefits to aid contraceptive counseling. This contrasts with the availability of information on contraceptive effectiveness11 and safety.12,13

To begin to address this gap, the U.S. Office of Population Affairs initiated a systematic review on the side effects and noncontraceptive benefits of intramuscular and subcutaneous depot medroxyprogesterone acetate (DMPA) injectable contraceptives. Side effects and health benefits of DMPA are particularly salient, because 73% of ever-users who discontinued DMPA reported doing so because of side effects.14 Two key questions guided this review: 1) What side effects are associated with progestin-only injectable contraceptive use? 2) What health benefits are associated with progestin-only injectable contraceptive use?

Back to Top | Article Outline

SOURCES

This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and focused on DMPA formulations currently available in the United States: 150 mg intramuscular and 104 mg subcutaneous DMPA.

We searched seven electronic bibliographic databases (PubMed, POPLINE, EMBASE, Web of Science, Campbell Collaboration Library of Systematic Reviews, the Cochrane Database of Systematic Reviews, and the Cochrane Center Register of Controlled Trials), using search phrases grouped by concepts related to the key questions (Appendix 1, available online at http://links.lww.com/AOG/B253). We also searched ClinicalTrials.gov using the search term “depot medroxyprogesterone.” Full-text articles were retrieved for studies not excluded after title and abstract review. Three Master's level analysts at Atlas Research LLC, including coauthor R.L.S., reviewed titles and abstracts of records and entered them into an electronic database. An independent reviewer at Atlas conducted a spot check (5%) of all studies excluded and validated 100% of the spot checked exclusions. Finally, the first author (S.D.) reviewed all excluded titles and verified the resulting database's abstractions against the included articles.

Back to Top | Article Outline

STUDY SELECTION

Studies were eligible for inclusion if they were reported in the English language and available as a full-text article. Inclusion and exclusion criteria (Appendix 2, available online at http://links.lww.com/AOG/B253) identified studies that could adequately answer the key questions. Included studies were required to have: 1) examined either key question 1 or 2 among nonbreastfeeding, healthy (average-risk health, which includes overweight and obese) females aged 13–49 years at risk of unintended pregnancy; 2) been conducted in clinical or academic settings; 3) had at least a two-group design, with at least one group using DMPA and at least one group using other contraception currently available in the United States; 4) controlled for confounding (eg, covariate adjustment, restriction, stratification, randomized group assignment), and 5) been published between January 1, 1985, and November 30, 2016.

The database search identified 4,004 references, and the search of ClinicalTrials.gov yielded 127 records of completed studies (Appendix 3, available online at http://links.lww.com/AOG/B253). Between both searches, 1,634 records were excluded owing to either duplicate citations (n=1,624) or dates outside of the predetermined timeframe (n=10). After all the exclusion criteria were applied, 80 articles remained and were retrieved for full-text review. Four of these articles were unavailable in full text. Seventy-six articles were reviewed in full, of which 26 articles were included in the systematic review.

Of the 50 articles that were excluded, six were systematic review articles; a hand-search of their references identified no unique articles that fit our review's inclusion criteria. One article was initially excluded because it included participants aged 12–18,15 which was 1 year outside of the inclusion criteria of 13–45 years; this article otherwise met inclusion criteria and was subsequently included, given the small age difference and our desire to err on the side of including more information.

Studies that met inclusion criteria were abstracted and categorized by study design. Study quality was assessed for internal and external validity, per established guidelines.16,17 Internal validity was assessed at the outcome level by rating the study as having low, moderate, or high risk of bias. External validity was assessed as good, fair, or poor, in relation to the population of average-risk, nonbreastfeeding reproductive-aged people in the United States. Descriptive summary measures were extracted. The review protocol is available on request.

This review defined “side effects” based on the definition provided in the World Health Organization's (WHO) family planning handbook.18 We adapted the WHO definition for the specific context of contraceptive methods as follows:

Any common (affecting at least 5% of people) noncontraceptive effect of a contraceptive pharmaceutical product without serious health effects, occurring at doses normally used in healthy women of reproductive age, that is either related to the pharmacologic properties of the drug or the physical attributes of the drug delivery system, and is a physical or mental feature apparent to the user.

We allowed that a side effect (eg, amenorrhea) could be perceived as positive or negative by the user, as its assessment can be patient- and circumstance-dependent.

Side effects listed for any type of contraceptive method in the 2011 WHO handbook (Appendix 4, available online at http://links.lww.com/AOG/B253) were included in this systematic review as search terms, with the exception of side effects related to transmission or acquisition of sexually transmitted infections, administration of method, and side effects not recognizable by patients. We also added search terms for nonserious nonrare health conditions (eg, yeast infection, urinary tract infection, and bacterial vaginosis) and common health-related concerns, such as reduction in sexual pleasure and delay in return to fertility once a contraceptive method is discontinued.3,19,20 Search terms were cross-referenced for completeness with side effects listed on a U.S. Food and Drug Administration birth control chart and in a 2018 update to the WHO handbook.

To define “health benefits,” we modified the WHO definition of “side effect” as follows:

Any reduction of a noncontraceptive serious health condition associated with a contraceptive pharmaceutical product occurring at doses normally used in healthy women of reproductive age that is either related to the pharmacologic properties of the drug or the physical attributes of the drug delivery system.

We distinguished beneficial side effects from health benefits in that the latter were uncommon (occurred in less than 5% of users or nonusers) and protected against a serious health condition.

Back to Top | Article Outline

RESULTS

This systematic review includes 26 articles describing 24 studies. There were zero randomized controlled trials, 13 prospective cohort studies, five retrospective cohort studies, four case–control studies, and two cross-sectional studies (Figure 1). The most frequently studied side effect was weight gain, reported in 13 of the 24 studies. Other reported outcomes were body composition (five studies), depressed or negative mood (three studies), mood swings (two studies), low libido or sexual interest (four studies), cancer protection (three studies), and other outcomes reported only once (five studies). Most of the studies (n=17) compared outcomes among DMPA users to nonhormonal method users (Figure 2). Eleven studies had moderate risk of bias and 13 had high risk of bias (Appendix 5, available online at http://links.lww.com/AOG/B253).

Fig 1.

Fig 1.

Fig. 2.

Fig. 2.

All included studies either used the intramuscular DMPA formulation or did not specify the type of DMPA used; unspecified DMPA was assumed to be the intramuscular formulation because this was the first formulation of DMPA developed.

The 24 studies included 61 comparisons between DMPA users and users of other methods. Forty-two of the 61 comparisons were between DMPA users and women using nonhormonal controls, such as the copper intrauterine device (Cu-IUD), or women using any type of nonhormonal method (withdrawal, barrier, fertility awareness-based, abstinence, and sometimes including Cu-IUD). Ten of the 61 comparisons included women using combined hormonal methods, referring to combined estrogen and progestin-containing pills, patches, or vaginal rings. Five comparisons included women using a mix of non-DMPA methods, and four comparisons included women using other progestin-only methods (implant or levonorgestrel intrauterine system) (Figure 2).

For the internal validity quality assessment, risk of bias was rated as “low” for zero of 61 comparisons, “moderate” for 19 of 61, and “high” for 42 of 61 comparisons (Appendix 5, http://links.lww.com/AOG/B253). The most common reasons for rating a study as having a high risk of bias were not controlling for major confounding variables or having high (greater than 50%) overall loss to follow-up, high differential loss to follow-up by contraceptive method group, or high risk of selection bias. Reasons for rating a study as having a moderate risk of bias included having loss to follow-up between 20% and 50%, not controlling for minor confounding variables, or having potential recall bias (Appendices 6–8, available online at http://links.lww.com/AOG/B253).

For external validity quality assessment, generalizability was rated as “good” in ten studies, “fair” in 14 studies, and “poor” in zero studies (Appendices 6–8, http://links.lww.com/AOG/B253). Two factors downgraded the study to “fair”: 1) either intentionally21–23 or unintentionally15,24–28 excluding women of obese body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) or 2) being conducted in a country other than the United States21–23,25–35 or from a small subpopulation in the United States,36 making the study population less generalizable to the U.S. primary care population.

Sixteen articles describing fourteen studies reported on weight- or body composition-related outcomes among DMPA injectable contraceptive users compared with their comparison group. There were nine prospective cohort studies and five retrospective cohort studies (Figure 1). Most of these studies reported either weight change, change in BMI, or both. Five studies15,21,24,29,37 measured body composition with the widely used dual-energy X-ray absorptiometry technique and reported on changes in total body or central fat mass; two studies reported on perceived or self-reported weight change.38,39 All but one study36 compared outcomes of DMPA-users to outcomes of nonhormonal method users (Figure 2). Six of the studies had moderate risk of bias, and eight had high risk of bias (Appendix 5, http://links.lww.com/AOG/B253). The articles by Berenson et al38 and Berenson and Rahman37 are from the same study; Berenson et al present the patient-reported subjective data, and Berenson and Rahman present the objective measurements. In this review, we referenced these as one study with multiple outcome measures. The articles by Dos Santos from 2014 to 2016 had partially overlapping study participants and different rates lost to follow-up, so we referenced them as two separate studies.21,22

Among the six studies with moderate risk of bias, five reported objective weight or BMI outcomes and three reported on body composition. Risk of bias was rated as “moderate” most frequently for attrition greater than 20% or not controlling for potential confounding by activity level or diet (Appendix 6, http://links.lww.com/AOG/B253); these are two lifestyle variables that can correspond to both contraceptive method choice and to weight gain. Four studies15,23–25 found DMPA users had statistically significant increases in weight, BMI, or body fat mass compared with Cu-IUD or other nonhormonal method users, and two studies21,22 found no statistically significant difference in change in BMI or body fat between DMPA and Cu-IUD users. The study by Batista et al found that DMPA users and Cu-IUD users had similar mean weight at baseline (62 kg vs 61 kg, respectively, P=.54), but at 1-year follow-up, DMPA users had statistically significantly higher mean weight than Cu-IUD users (DMPA 65 kg vs Cu-IUD 62 kg, P<.001).23 Clark et al found that new DMPA users had gained approximately 2.6 kg at 1 year, whereas nonhormonal method users had gained less than 0.1 kg (P<.03)24; at 2.5 years, DMPA users had gained a total mean of 6.1 kg, whereas nonhormonal method users gained, as reported in the article, “virtually no weight” (P<.03). Modesto et al25 found that new DMPA users gained 1.3 kg over 1 year, whereas Cu-IUD users gained 0.2 kg (P<.001). Modesto's 10-year follow-up found that DMPA users who had remained in the study and continued using DMPA for the entire period had gained 6.6 kg, whereas Cu-IUD users had gained 4.9 kg (P=.04); however, the 10-year data are at high risk of bias owing to a 79–90% loss to follow-up.25 Dos Santos et al found no statistically significant differences in change of BMI at 1 year between DMPA and Cu-IUD users.21,22

With respect to body composition, Bonny et al found that DMPA users had a total body fat increase of 10.3% over 6 months, whereas nonhormonal method users had a total body fat decrease of 0.7% (P=.04).15 Clark et al similarly found a statistically significant increase in fat mass, along with an increase in central-to-peripheral fat mass ratio, in DMPA users over 2.5 years compared with no change in nonhormonal method users; duration of DMPA use was the strongest predictor of these outcomes.24 Dos Santos et al found no statistically significant difference between groups in total body fat over 1 year.21

Of the eight studies examining weight-related outcomes that had high risk of bias, seven reported on objective weight or BMI measurements, two reported on subjective or perceived weight gain, and two additionally reported on body composition. Among the studies reporting objective measurements, three found that DMPA users had statistically significantly greater increases in weight or BMI from baseline compared with their comparison group, in follow-up durations ranging from 2 to 5 years.26,36,37 Three studies found equivalent increases from baseline between groups, in follow-up durations ranging from 1 to 10 years.27–29,40 One study found outcomes differed by baseline BMI: DMPA users in the baseline normal- and overweight-BMI groups gained more weight over 3 years than Cu-IUD users, but differences in weight gain were not found between DMPA and Cu-IUD users in those who had baseline obese BMI.30

Among the two studies that measured subjective weight gain, both found a statistically significantly greater perceived weight gain in DMPA users compared with nonhormonal method users, with Nault et al reporting a 75% sensitivity and 84% specificity for true weight gain.38,39

The study of body composition by Dal'Ava et al29 found that DMPA users had a statistically significant increase in central body fat compared with Cu-IUD users (P=.04) over 1 year. Similarly, Berenson and Rahman37 found that DMPA users had increased central-to-peripheral fat ratio compared with nonhormonal method users over a 3-year follow-up (P<.01). For total body fat measurements, Dal'Ava et al found no statistically significant difference between groups at 1 year, whereas Berenson and Rahman found that DMPA users had statistically significantly increased fat mass compared with nonhormonal method users over a 3-year follow-up (P<.01).29,37

Notably, six of the studies with high risk of bias received their bias rating because of potential selection bias. For example, the studies by Pantoja et al,30 Bahamondes et al,26 and Taneepanichskul et al27,28 retrospectively sampled only the DMPA and other contraceptive method users who had used their method continuously for at least 3, 5, and 10 years, respectively. Excluding users who discontinued their method earlier, which for DMPA users could have been due to weight gain, could have resulted in an underestimate of true weight gain in the DMPA user group (Appendix 6, http://links.lww.com/AOG/B253).

Four studies (Berenson et al,38 Civic et al,41 Brown et al,42 and Ott et al43) reported on depressed or negative mood, nervousness, “well-being,” or mood swings in DMPA users compared with control groups. All four studies were prospective cohort studies rated as having a high risk of bias for inadequately controlling for major confounding variables such as relationship status or relationship satisfaction, life stressors, or financial or socioeconomic instability; the study by Civic et al41 additionally had a risk of selection bias owing to a 60% loss to follow-up or method discontinuation over 3 years (Appendix 7, http://links.lww.com/AOG/B253). Comparisons were made between DMPA and nonhormonal methods, combined hormonal contraceptives, and a mixed group of non-DMPA methods. Berenson et al38 and Civic et al41 used standardized measurement instruments, whereas Brown et al42 and Ott et al43 used investigator-designed instruments that had not been validated.

The study by Berenson et al38 followed new contraceptive method users for 2 years and found no statistically significant difference in nervousness or depressive symptoms, comparing users of DMPA to users of nonhormonal methods. In contrast, Civic et al41 found that DMPA users had 44% greater adjusted odds of having increased depressive symptoms compared with nonusers of DMPA (as measured by a difference of 10 or more points on the Community Epidemiology Survey—Depression Scale) (P=.047; 95% CI 1.00–2.07). Brown et al42 found that both DMPA and combined oral contraceptive users had statistically significant greater overall “negative well-being” than nonhormonal method users, measured over three menstrual cycles. Ott et al43 found that DMPA users compared with combined oral contraceptive and other non-DMPA users had more “negative mood,” with statistical significance that had unclear clinical significance (maximum absolute difference 0.22 points on an index range of 3–15 points).

Among the two studies that reported on mood swings for DMPA users compared with nonhormonal method users, Berenson et al38 found DMPA users had decreased odds of mood swings (adjusted odds ratio [aOR]=0.66; 95% CI 0.45–0.97), whereas Brown et al42 found no statistically significant difference.

Four studies examined low sexual interest or loss of libido in DMPA users compared with control groups; two of these were prospective cohort studies (Berenson et al38 and Ott et al43) and two were cross-sectional studies (Boozalis et al44 and Schaffir et al45). Boozalis et al44 was rated as having a moderate risk of bias for not controlling for potential confounding by relationship satisfaction; the studies by Berenson et al38 and Ott et al43 were rated as having a high risk of bias for additionally not controlling for potential confounding by relationship status and life stressors. Berenson et al38 also had a high risk of bias owing to a high loss to follow-up or discontinuation of 55–61% at 2 years (no data were reported for their 1-year follow-up when attrition was less than 50%). The study by Schaffir et al45 was rated as having a high risk of bias because of potential selection bias, because participants had to have been using their method for an average of 4–5 years without an interest in changing methods (Appendix 7, http://links.lww.com/AOG/B253).

Boozalis et al44 conducted a follow-up phone interview at 6 months of method use among participants from a large prospective cohort study. They found DMPA users had greater odds of reporting lack of interest in having sex for several months in the previous 6 months compared with Cu-IUD users (aOR=2.61; 95% CI 1.47–4.61). A subgroup of these participants additionally had baseline interviews from the time of method initiation. In this group, the 6-month estimates were adjusted for lack of interest in having sex at baseline, with the resulting aOR no longer statistically significant (aOR=1.99; 95% CI 0.79–5.05), suggesting that the initial findings could be attributed to lower baseline interest in sex among the DMPA initiators.44

Berenson et al38 used a symptom checklist to assess loss of libido and dyspareunia in DMPA, combined oral contraceptive, and nonhormonal method users over 2 years. Users of DMPA had greater odds of loss of libido compared with nonhormonal method users (aOR=2.24; 95% CI 1.50–3.34), whereas combined oral contraceptive users had similar odds compared with nonhormonal method users (aOR=0.84; 95% CI 0.56–1.26).38 There was no statistically significant difference in odds of dyspareunia in DMPA users compared with nonhormonal method users.

Ott et al43 used a 5-point Likert-type scale to ask participants daily whether they were interested in having sex. Weekly mean sexual interest scores were not statistically significantly different between DMPA and combined oral contraceptive users.

Finally, Schaffir et al45 used the validated Female Sexual Function Index questionnaire in a cross-sectional analysis and found that DMPA users and combined oral contraceptive users had no statistically significant difference in scores of desire (index range 0–6: DMPA 3.8 vs combined oral contraceptive 4.2, P=.27), arousal (index range 0–6: DMPA 4.8 vs combined oral contraceptive 5.0, P=.46), and total Female Sexual Function Index scores (index range 0–36: DMPA 28.8 vs combined oral contraceptive 30.1, P=.28).

The single study reporting on menstrual symptoms was a prospective cohort study (Berenson et al38) that used a symptom checklist at visits every 6 months for 2 years. It was at high risk of bias owing to loss to follow-up (55–61% attrition at 2 years). At 2 years, DMPA compared with nonhormonal method users had greater odds of reporting continuous bleeding longer than 20 days (aOR=13.37; 95% CI 5.35–33.38), greater odds of reporting intermenstrual bleeding (aOR=3.61; 95% CI 2.22–5.90), greater odds of reporting missed periods (aOR=96.90; 95% CI 53.81–174.47), and lower odds of reporting cramping (aOR=0.35; 95% CI 0.24–0.51).

Various other potential side effects were assessed by single studies. In Berenson et al's38 prospective cohort study, DMPA users had statistically significant lower odds of bloating (aOR=0.48; 95% CI 0.34–0.69) and greater odds of loss of energy (aOR=1.55; 95% CI 1.10–2.18), but equivalent odds of mastalgia, acne, scalp hair loss, hirsutism, and headache compared with nonhormonal method users. Ziaei et al's31 prospective cohort study compared new DMPA users to withdrawal users and found that urinary tract infection cumulative incidence over a 3-month period was 5% compared with 0.5%, respectively (P=.018). Users of DMPA were also statistically more likely to experience urologic symptoms (P<.001), including urinary frequency (22% in DMPA users vs 1.5% in withdrawal users) and urinary incontinence (7.5% in DMPA users vs 1% in withdrawal users). This study was at moderate risk of bias: although it matched participants on age, gravidity, and socioeconomic and educational status at baseline, it did not account for potential confounding by frequency of sexual activity (Appendix 8, http://links.lww.com/AOG/B253).

Three case–control studies examined whether DMPA protected against cancers; two studies were rated as having a moderate risk of bias (Cuevas et al,33 Wilailak et al35), and one was rated as having a high risk of bias (Silpisornkosol et al34). Notably, the search terms for cancer were broad to include all cancer types, but the studies that resulted and ultimately were included were of genitourinary tract cancers only.

Cuevas et al33 and Wilailak et al35 examined history of DMPA use among patients hospitalized for treatment of epithelial ovarian cancer and those in control groups hospitalized in a similar timeframe with nongynecologic diagnoses, matched on age, hospital, and year of interview. Cuevas et al33 found no difference in odds of ovarian cancer among those with history of DMPA use compared with never-users (aOR=1.07; 95% CI 0.6–1.8). Wilailak et al,35 however, found that those with history of DMPA use had 39% lower odds of epithelial ovarian cancer compared with never-users (aOR=0.61; 95% CI 0.44–0.85). Odds of epithelial ovarian cancer were even lower in those with history of DMPA use for more than 3 years compared with never-users (aOR=0.17; 95% CI 0.07–0.39).

Silpisornkosol et al34 used a similar matched case–control study design and examined history of DMPA use in endometrial cancer patients compared with those in a control group. They found that history of DMPA use was associated with 79% lower odds of endometrial cancer compared with never-users (aOR=0.21; 95% CI 0.06–0.79). Notably, they excluded women who had first used DMPA in the year before the cancer diagnosis, as they were thought likely to have been prescribed DMPA for symptoms related to the yet-undiagnosed endometrial cancer.

All three case–control studies collected DMPA use history retrospectively and were therefore subject to recall bias, as gynecologic cancer patients may report previous exposure to hormones differently than patients without a history of gynecologic cancer. An additional weakness of the Silpisornkosol et al34 study is that they did not account for potential confounding by history of obesity, which is an important risk factor for endometrial cancer and might also be related to DMPA use (Appendix 8, http://links.lww.com/AOG/B253).

Bahamondes et al's32 case–control study examined the effects of DMPA on tubal infertility. This study retrospectively collected data on history of DMPA use. It was rated as having a moderate risk of bias owing to potential recall bias. Those with a history of DMPA use had lower odds of infertility caused by tubal obstruction compared with never-users of DMPA (aOR=0.35; 95% CI 0.1–0.8).32 Barrier contraceptive users and combined oral contraceptive users had even lower odds (aOR=0.16; 95% CI 0.1–0.5; aOR=0.27; 95% CI 0.2–0.3, respectively) compared with never-users of barrier and combined oral contraceptive methods. When considering only those with secondary tubal infertility (ie, tubal infertility after having had at least one live birth), the protective effects of DMPA and barrier method use were no longer statistically significant whereas the protective effect of combined oral contraceptives persisted, suggesting the protective effects of DMPA may apply only to primary tubal infertility (ie, tubal infertility in those who have never had a live birth).

Back to Top | Article Outline

DISCUSSION

We synthesized the evidence on DMPA-related side effects and health benefits to improve the quality of contraceptive counseling.3,19 Our findings suggest an association between DMPA use and weight gain, including increased body fat mass, and support an association with irregular bleeding and amenorrhea. Inconsistent evidence exists for an association between DMPA use and mood or libido changes. Limited evidence suggests an association between DMPA use and decreased risk of cancers and tubal infertility.

Our systematic review found no eligible randomized controlled trials. Although randomizing women to contraceptive methods can present ethical and logistical challenges,46 it may be possible in some cases47,48 and would enhance the quality of evidence. Notably, the prospective cohort studies included in this review had a moderate or high risk of bias, highlighting the need for improved methodology for nonrandomized studies. Providers should recognize that the lack of data indicating the statistical significance of a specific side effect does not preclude individual patients from experiencing that side effect. Conversely, recognized side effects are not experienced universally, and some identified side effects could be a result of residual confounding. By being transparent about the evidence's complexity and limitations, providers can build trust with patients and support their contraceptive decision making in a patient-centered and evidence-based manner.

The existing evidence on weight and body composition has moderate to high risk of bias but suggests weight gain ranging from 1.3 to 3 kg among DMPA users over 1 year, compared with 0.1–1.1 kg weight gain in the nonhormonal method comparison groups.23–25 Likewise, there are statistically significantly greater increases in total body or central fat mass among DMPA users compared nonhormonal method users.15,21,24,29,37 A 2016 Cochrane review similarly concluded that the quality of evidence regarding weight is low; they reported a possible increase in weight gain of less than 2 kg at 6 or 12 months in progestin-only users compared with nonhormonal users.49 The Cochrane review excluded24,37–39 or did not consider23,26 five studies that we included, which unanimously showed a statistically significant increase in weight gain in DMPA users compared with their nonuser counterparts.

Whether weight gain differs between baseline obese and nonobese DMPA users was addressed in two of our review's studies, both of which found greater weight gain in baseline nonobese DMPA users.30,37 Given the heterogeneity and limitations of these findings, higher-quality studies are needed to better understand the relationship between DMPA use and weight gain, and the influence of baseline weight.

Our findings were inconclusive on whether DMPA use affects mood, sexual interest, and libido. Studies were limited by inability to address important potential confounders such as life stressors, relationship status, and relationship satisfaction.50,51 Two studies by Skovlund et al, published after our publication timeframe, found positive associations between hormonal contraceptive methods and, assessed separately, risk of depression and suicide, but findings were subject to residual confounding.52,53

A recent review by Worly et al,54 which included descriptive studies and those without a comparison group, found that concerns for depression in users of DMPA were unfounded, except possibly in postpartum women. Future prospective studies with appropriate comparison groups, validated instruments, and attention to important confounding variables are needed to assess this important aspect of contraceptive users' lives.55

Few studies evaluated menstrual changes, despite it being one of the most prominently perceived side effects of DMPA.18 Findings of prolonged bleeding, amenorrhea, spotting, and decreases in cramping were consistent with commonly-cited side effects of DMPA.18 Women may perceive menstrual side effects positively or negatively, and these perceptions should be explored in contraceptive counseling.

Few studies evaluated the protective effect of DMPA use on cancer risk. The included case–control studies suggest that DMPA may reduce risk of endometrial and ovarian cancers. A recent retrospective cohort study by Mørch et al,56 published after our publication timeframe, found a positive association between hormonal contraceptive methods and breast cancer, but found no association for DMPA users specifically.

Other potential side effects, such as acne, scalp hair loss, headache, hirsutism, loss of energy, and mastalgia, may be important to women's contraceptive decision-making and are understudied areas. Furthermore, none of the eligible studies evaluated return to fertility after discontinuing DMPA. A 2011 review on return to fertility similarly found no DMPA studies that met their inclusion criteria.57

Patient-centered counseling should incorporate the available evidence while acknowledging its limitations and recognizing the value of women's lived experience of contraceptive use.

Back to Top | Article Outline

REFERENCES

1. Castle S, Askew I. Contraceptive discontinuation: reasons, challenges, and solutions. Available at: http://www.familyplanning2020.org/microsite/contraceptive-discontinuation. Retrieved February 15, 2018.
2. Madden T, Secura GM, Nease RF, Politi MC, Peipert JF. The role of contraceptive attributes in women's contraceptive decision making. Am J Obstet Gynecol 2015;213:46.e1–6.
3. Lessard LN, Karasek D, Ma S, Darney P, Deardorff J, Lahiff M, et al. Contraceptive features preferred by women at high risk of unintended pregnancy. Perspect Sex Reprod Health 2012;44:194–200.
4. Backman T, Huhtala S, Luoto R, Tuominen J, Rauramo I, Koskenvuo M. Advance information improves user satisfaction with the levonorgestrel intrauterine system. Obstet Gynecol 2002;99:608–13.
5. Lei ZW, Chun Wu S, Garceau RJ, Jiang S, Yang QZ, Wang WL, et al. Effect of pretreatment counseling on discontinuation rates in Chinese women given depo-medroxyprogesterone acetate for contraception. Contraception 1996;53:357–61.
6. Canto De Cetina TE, Canto P, Ordoñez Luna M. Effect of counseling to improve compliance in Mexican women receiving depot-medroxyprogesterone acetate. Contraception 2001;63:143–6.
7. Providing quality family planning services: recommendations of CDC and the U.S. Office of Population Affairs. Available at: https://www.cdc.gov/mmwr/preview/mmwrhtml/rr6304a1.htm?s_cid=rr6304a1_w. Retrieved February 14, 2018.
8. Guendelman S, Denny C, Mauldon J, Chetkovich C. Perceptions of hormonal contraceptive safety and side effects among low-income Latina and non-Latina women. Matern Child Health J 2000;4:233–9.
9. Dehlendorf C, Levy K, Kelley A, Grumbach K, Steinauer J. Women's preferences for contraceptive counseling and decision making. Contraception 2013;88:250–6.
10. Yee LM, Simon MA. Perceptions of coercion, discrimination and other negative experiences in postpartum contraceptive counseling for low-income minority women. J Health Care Poor Underserved 2011;22:1387–400.
11. Sundaram A, Vaughan B, Kost K, Bankole A, Finer L, Singh S, et al. Contraceptive failure in the United States: estimates from the 2006–2010 national Survey of family growth. Perspect Sex Reprod Health 2017;49:7–16.
12. Center for Drug Evaluation and Research. New drug application: medroxyprogesterone acetate injectable suspension. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2004/020246Orig1s025.pdf. Retrieved May 14, 2018.
13. Curtis KM. U.S. medical eligibility criteria for contraceptive use, 2016. MMWR Recomm Rep 2016;65:1–103.
14. Centers for Disease Control and Prevention. Key statistics from the National Survey of Family Growth—I listing. Available at: https://www.cdc.gov/nchs/nsfg/key_statistics/i.htm. Retrieved February 14, 2018.
15. Bonny AE, Secic M, Cromer BA. A longitudinal comparison of body composition changes in adolescent girls receiving hormonal contraception. J Adolesc Health 2009;45:423–5.
16. Tregear SJ, Gavin LE, Williams JR. Systematic review evidence methodology: providing quality family planning Services. Am J Prev Med 2015;49(2 suppl 1):S23–30.
17. U.S. Preventive Services Task Force. Procedure manual. Available at: https://www.uspreventiveservicestaskforce.org/Page/Name/procedure-manual. Retrieved February 15, 2018.
18. World Health Organization. Family planning: a global handbook for providers. Retrieved 2011 edition on February 15, 2018. Updated 2018 edition available at: https://www.who.int/reproductivehealth/publications/fp-global-handbook/en/. Retrieved December 31, 2018.
19. Jackson AV, Karasek D, Dehlendorf C, Foster DG. Racial and ethnic differences in women's preferences for features of contraceptive methods. Contraception 2016;93:406–11.
20. Coombe J, Harris ML, Loxton D. What qualities of long-acting reversible contraception do women perceive as desirable or undesirable? A systematic review. Sex Health 2016 [Epub ahead of print].
21. dos Santos Pde N, Modesto WO, Dal'Ava N, Bahamondes MV, Pavin EJ, Fernandes A. Body composition and weight gain in new users of the three-monthly injectable contraceptive, depot-medroxyprogesterone acetate, after 12 months of follow-up. Eur J Contracept Reprod Health Care 2014;19:432–8.
22. Silva dos Santos PdeN, de Souza AL, Batista GA, Melhado-Kimura V, de Lima GA, Bahamondes L, et al. Binge eating and biochemical markers of appetite in new users of the contraceptive depot medroxyprogesterone acetate. Arch Gynecol Obstet 2016;294:1331–6.
23. Batista GA, Souza ALde, Marin DM, Sider M, Melhado VC, Fernandes AM, et al. Body composition, resting energy expenditure and inflammatory markers: impact in users of depot medroxyprogesterone acetate after 12 months follow-up. Arch Endocrinol Metab 2017;61:70–5.
24. Clark MK, Dillon JS, Sowers M, Nichols S. Weight, fat mass and central distribution of fat increase when women use depot-medroxyprogesterone acetate for contraception. Int J Obes 2005;29:1252–8.
25. Modesto W, de Nazaré Silva dos Santos P, Correia VM, Borges L, Bahamondes L. Weight variation in users of depot-medroxyprogesterone acetate, the levonorgestrel-releasing intrauterine system and a copper intrauterine device for up to ten years of use. Eur J Contracept Reprod Health Care 2015;20:57–63.
26. Bahamondes L, Del Castillo S, Tabares G, Arce XE, Perrotti M, Petta C. Comparison of weight increase in users of depot medroxyprogesterone acetate and copper IUD up to 5 years. Contraception 2001;64:223–5.
27. Taneepanichskul S, Reinprayoon D, Jaisamrarn U. Effects of DMPA on weight and blood pressure in long term acceptors. Contraception 1999;59:301–3.
28. Taneepanichskul S, Reinprayoon D, Khaosaad P. Comparative study of weight change between long-term DMPA and IUD acceptors. Contraception 1998;58:149–51.
29. Dal'Ava N, Bahamondes L, Bahamondes MV, Bottura BF, Monteiro I. Body weight and body composition of depot medroxyprogesterone acetate users. Contraception 2014;90:182–7.
30. Pantoja M, Medeiros T, Baccarin MC, Morais SS, Bahamondes L, dos Santos Fernandes AM. Variations in body mass index of users of depot-medroxyprogesterone acetate as a contraceptive. Contraception 2010;81:107–11.
31. Ziaei S, Ninavaei M, Faghihzadeh S. Urinary tract infection in the users of depot-medroxyprogesterone acetate. Acta Obstet Gynecol Scand 2004;83:909–11.
32. Bahamondes L, Bueno JGR, Hardy E, Vera S, Pimentel E, Ramos M. Identification of main risk factors for tubal infertility. Fertil Steril 1994;61:478–82.
33. Cuevas HR, Salazar SB, Palet A, Ontiveros P, Slipisornkosol S, Pardthaisong T, et al. Depot-medroxyprogesterone acetate (DMPA) and risk of epithelial ovarian cancer. Int J Cancer 1991;49:186–90.
34. Silpisornkosol S, Sahapong V, Theetranont C, Boosiri B, Chutivongse S, Wongsrchanalai C, et al. Depot-medroxyprogesterone acetate (dmpa) and risk of edometrial cancer. Int J Cancer 1991;49:186–90.
35. Wilailak S, Vipupinyo C, Suraseranivong V, Chotivanich K, Kietpeerakool C, Tanapat Y, et al. Depot medroxyprogesterone acetate and epithelial ovarian cancer: a multicentre case-control study: DMPA and epithelial ovarian cancer. BJOG 2012;119:672–7.
36. Espey E, Steinhart J, Ogburn T, Qualls C. Depo-provera associated with weight gain in Navajo women. Contraception 2000;62:55–8.
37. Berenson AB, Rahman M. Changes in weight, total fat, percent body fat, and central-to-peripheral fat ratio associated with injectable and oral contraceptive use. Am J Obstet Gynecol 2009;200:329.e1–8.
38. Berenson AB, Odom SD, Breitkopf CR, Rahman M. Physiologic and psychologic symptoms associated with use of injectable contraception and 20 μg oral contraceptive pills. Am J Obstet Gynecol 2008;199:351.e1–12.
39. Nault AM, Peipert JF, Zhao Q, Madden T, Secura GM. Validity of perceived weight gain in women using long-acting reversible contraception and depot medroxyprogesterone acetate. Am J Obstet Gynecol 2013;208:48.e1–8.
40. Vickery Z, Madden T, Zhao Q, Secura GM, Allsworth JE, Peipert JF. Weight change at 12 months in users of three progestin-only contraceptive methods. Contraception 2013;88:503–8.
41. Civic D, Scholes D, Ichikawa L, LaCroix AZ, Yoshida CK, Ott SM, et al. Depressive symptoms in users and non-users of depot medroxyprogesterone acetate. Contraception 2000;61:385–90.
42. Brown SG, Morrison LA, Larkspur LM, Marsh AL, Nicolaisen N. Well-being, sleep, exercise patterns, and the menstrual cycle: a comparison of natural hormones, oral contraceptives and depo-provera. Women Health 2008;47:105–21.
43. Ott MA, Shew ML, Ofner S, Tu W, Fortenberry JD. The influence of hormonal contraception on mood and sexual interest among adolescents. Arch Sex Behav 2008;37:605–13.
44. Boozalis A, Tutlam NT, Chrisman Robbins C, Peipert JF. Sexual desire and hormonal contraception. Obstet Gynecol 2016;127:563–72.
45. Schaffir JA, Isley MM, Woodward M. Oral contraceptives vs injectable progestin in their effect on sexual behavior. Am J Obstet Gynecol 2010;203:545.e1–5.
46. Piazza A, Schwirian K, Scott F, Wilson MD, Zite NB, Creinin MD. Women's preferences for permanent contraception method and willingness to be randomized for a hypothetical trial. Contraception 2019;99:56–60.
47. Achilles SL, Chen BA, Lee JK, Gariepy AM, Creinin MD. Acceptability of randomization to levonorgestrel versus copper intrauterine device among women requesting IUD insertion for contraception. Contraception 2015;92:572–4.
48. Hubacher D, Spector H, Monteith C, Chen PL, Hart C. Rationale and enrollment results for a partially randomized patient preference trial to compare continuation rates of short-acting and long-acting reversible contraception. Contraception 2015;91:185–92.
49. Lopez LM, Ramesh S, Chen M, Edelman A, Otterness C, Trussell J, et al. Progestin-only contraceptives: effects on weight. The Cochrane Database of Systematic Reviews 2016, Issue 8. Art. No.: CD008815. doi: 10.1002/14651858.CD008815.pub4.
50. Harvey SM, Oakley LP, Washburn I, Agnew CR. Contraceptive method choice among young adults: influence of individual and relationship factors. J Sex Res 2018;55:1106–15.
51. Romans SE, Asllani E, Clarkson RF, Meiyappan S, Petrovic MJ, Tang D. Women's perceptions of influences on their mood. Women Health 2009;49:32–49.
52. Skovlund CW, Mørch LS, Kessing LV, Lidegaard Ø. Association of hormonal contraception with depression. JAMA Psychiatry 2016;73:1154–62.
53. Skovlund CW, Mørch LS, Kessing LV, Lange T, Lidegaard Ø. Association of hormonal contraception with suicide attempts and suicides. Am J Psychiatry 2017;175:336–42.
54. Worly BL, Gur TL, Schaffir J. The relationship between progestin hormonal contraception and depression: a systematic review. Contraception 2018;97:478–89.
55. Higgins JA, Smith NK. The sexual acceptability of contraception: reviewing the literature and building a new concept. J Sex Res 2016;53:417–56.
56. Mørch LS, Skovlund CW, Hannaford PC, Iversen L, Fielding S, Lidegaard Ø. Contemporary hormonal contraception and the risk of breast cancer. N Engl J Med 2017;377:2228–39.
57. Mansour D, Gemzell-Danielsson K, Inki P, Jensen JT. Fertility after discontinuation of contraception: a comprehensive review of the literature. Contraception 2011;84:465–77.

Supplemental Digital Content

Back to Top | Article Outline
© 2019 by the American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.