Contraception 91 (2015) 19 – 24
Original research article
What happens when we routinely give doxycycline to medical abortion patients?☆,☆☆ Laura J. Frye⁎, Erica Chong, Beverly Winikoff, NCT01799252 Trial Investigators 1 Gynuity Health Projects, New York, NY 10010, USA Received 1 July 2014; revised 2 September 2014; accepted 2 September 2014
Abstract Objectives: Routine provision of antibiotics following medical abortion is common yet practitioners and professional societies differ on its utility. Our study compares the side effects experienced by women who were prescribed doxycycline following medical abortion to those who were not and assesses the adherence to one prescribed regimen. Study design: This was a prospective, observational, open-label study from a convenience sample. Women seeking medical abortion were enrolled in nine study sites, including four clinics that routinely prescribe a seven-day course of doxycycline (Doxycycline arm) and five clinics that do not routinely prescribe any antibiotics (No Doxycycline arm). Seven to fourteen days following the administration of mifepristone, women were asked to self-administer a computer-based survey. The survey asked about side effects experienced (both arms) and adherence to the regimen (Doxycycline arm only). Results: Five hundred eighty-one women were enrolled (278 in the Doxycycline arm and 303 in the No Doxycycline arm). There was a trend toward increased nausea in the Doxycycline arm (47.8% vs. 40.9%; p=.056) and a statistically significant difference in vomiting (25.2% vs. 18.5%; p=.032). Almost all women in the Doxycycline arm reported taking at least one pill, however only 28.3% reported “perfect adherence.” The most common reasons reported for taking fewer pills than instructed were that participants were still taking them (beyond 7 days) or that they forgot to take them. Conclusion: Women who were prescribed doxycycline following medical abortion reported moderate adherence and experienced significantly more vomiting than their counterparts. Implications: In the absence of robust evidence that prescribing 7 days of doxycycline following medical abortion is effective at reducing serious infections, these data can assist the public health community with deciding whether routine provision is the most appropriate strategy. © 2014 Elsevier Inc. All rights reserved. Keywords: Antibiotics; Doxycycline; Mifepristone; Misoprostol; Infection; Compliance
1. Introduction Early first trimester medical abortion is effective [1,2], highly acceptable to women [3–7], and safe [2,8,9]. The risk of infection following medical abortion is small, at less than 1% [10], with the rate of serious infection even lower, at .06/1000 [11]. Between 2001 and 2005, five women who had medical abortions in North America were infected with Clostridium ☆
The authors declare no conflicts of interest. Clinical Trial Registration Number: NCT01799252. ⁎ Corresponding author at: Gynuity Health Projects, 15 East 26th Street, Suite 801, New York, NY 10010. Tel.: +1 212 448 1230; fax: +1 212 448 1260. E-mail address: [email protected] (L.J. Frye). 1 Dr. Carol Ball, Dr. Lisa Harris, Dr. E. Steve Lichtenberg, Dr. Jules Marsh, Dr. Tamer Middleton, Dr. Amitasrigowri Murthy, Dr. Linda Prine. ☆☆
http://dx.doi.org/10.1016/j.contraception.2014.09.001 0010-7824/© 2014 Elsevier Inc. All rights reserved.
sordellii bacteria which caused toxic shock-like illness and death [12–14]. Although these fatal infections were extremely rare and no link was established between medical abortion and clostridia infection, Planned Parenthood Federation of America (PPFA) was prompted by these deaths to change its medical abortion protocols in 2006 to replace vaginal with buccal administration and to require routine antibiotic coverage [15]. These changes were widely adopted across the United States even among many independent providers. A retrospective review of data from PPFA showed a decrease over time in the serious infection rate after medical abortion following these protocol changes [11]. However, a subsequent report suggests that finding may have been a period effect [16,17]. Clifford and Daley estimated from the earlier PPFA data that 2500 women would need to be treated
20
L.J. Frye et al. / Contraception 91 (2015) 19–24
with antibiotics for a week to prevent one serious infection, and concluded that antibiotic prophylaxis for medical abortion is not warranted [18]. Indeed, recommendations from the World Health Organization, the Society of Family Planning, the American College of Obstetricians and Gynecologists, the National Abortion Federation, and the United States Food and Drug Administration all do not support routine use of preventive antibiotics for medical abortion [19–23]. To be clear, there is robust evidence of the efficacy of routine antibiotic prophylaxis following surgical abortion; however no clinical trials provide the same evidence base for medical abortion. Despite the lack of conclusive evidence on the efficacy of routine antibiotic treatment following medical abortion, this practice is common [24]. Considering that over 85,000 medical abortions are performed in the US each year, the advantages and disadvantages of any associated routine practice are amplified. At the time the study was conducted, doxycycline was the most widely used antibiotic following medical abortion and was typically given in a seven-day treatment dose of 100 mg twice daily. This treatment does not come without costs financially, logistically, and in terms of side effects. Doxycycline is associated with gastrointestinal, skin, and urogenital side effects [25] and while allergy is rare, it can include hypersensitivity syndrome reaction and serum sickness-like reaction [26]. Difficulty adhering to any antibiotic regimen would have implications for the benefits of routine prescription. Extensive evidence suggests that patient compliance with antibiotics is low [27,28]. A study of patients treated with doxycycline for Chlamydia found that 16% of patients fully adhered (as measured by electronic pill bottle cap) and 67% of patients had at least one instance of 24 h between doses [29]. While beyond the scope of this paper, public health concerns around the overuse of antibiotics and the development of antimicrobial resistance are important when examining the pros and cons of routine provision [30]. In summary, robust evidence is lacking on the effectiveness of doxycycline treatment following medical abortion but most women in the US receive it. This study seeks to add to the literature evidence of the side effects women experience with post-abortion doxycycline treatment and their adherence to the regimen.
2. Materials and methods This was a non-randomized, observational study conducted from October 2012 to December 2013. Participants were recruited from 9 clinics in the United States. Four clinics routinely prescribed doxycycline following medical abortion and five did not. All sites scheduled follow up visits at seven days following administration of mifepristone and, in the doxycycline arm, all sites dispensed the doxycycline directly to participants who were counseled to start the doxycycline the day they took mifepristone. Sites prescribed
ancillary medicines (analgesics, antiemetics, etc.) according to their regular protocols or patient needs. Women presenting for a medical abortion in the study clinics who could read English or Spanish were eligible for the study. Any woman who was currently taking antibiotics for reasons unrelated to her medical abortion was excluded as was any woman who had previously participated in this study. On the day they took their mifepristone in the clinic, eligible women were provided with a study information card which listed a unique Study ID number and a website where they could access a computer-based questionnaire 7–14 days later. Women could take the questionnaire either at the clinic during their follow-up visits, or off-site wherever they had an internet connection. If participants chose to take the questionnaire at the follow-up visit, they were given a laptop to self-administer it in a private location. Women who did not return for follow-up were called and reminded of the possibility of participating in the study using the website and Study ID provided to them. A woman was considered to be enrolled in the study once she accessed the website and gave her informed consent. The questionnaire used DatStat Illume v. 4.11 software and took 5–10 minutes to complete. The survey contained questions on demographics, medications dispensed or prescribed, side effects experienced and adherence to the doxycycline regimen (Doxycycline arm only). All questions related to the time period between taking the mifepristone in the clinic and the date of survey administration (7–14 days later). Women were provided with a list of side effects including: nausea, headache, vomiting, diarrhea, fever/chills, allergic reaction (rash, itching, wheezing) and yeast infection. They were asked to check each side effect experienced and had the option of writing in any others. We did not consider pain or bleeding to be side effects as they are the main effects of treatment. For each side effect women reported they were asked the severity, duration, and whether they took medications to treat it. In the Doxycycline arm, adherence was measured by asking the dates when women started and stopped taking doxycycline, if any doses were missed, if two consecutive doses were missed, and the number of pills remaining at the time of survey completion. All women who took fewer pills than prescribed were asked to provide a reason. “Perfect adherence” was defined as not missing any doses, having 0 or 1 pill remaining at the follow-up visit, and taking the medicine for 7 or 8 days. Women were also asked about elements of antibiotic counseling they received. No information on infection rates was obtained as the goals of the study were to focus on adherence and side effects of a commonly used antibiotic regimen. A sample size of 610 women (305 in each group) would be sufficient for a one-sided test to detect with 80% power a 10% lower nausea rate in the No Doxycycline arm (assuming a 50% nausea rate in the Doxycycline arm) [31]. A sample size of 305 in the Doxycycline arm would also allow an estimation of
L.J. Frye et al. / Contraception 91 (2015) 19–24
21
Fig. 1. Enrollment flow diagram.
adherence proportions with precisions of ± 6 percentage points. During the trial period there were two national doxycycline shortages which halted recruitment in the Doxycycline arm and limited the sample size. Statistical Package for Social Scientists (SPSS) v 19 and STATA v 12 were used for statistical analyses. Fisher's exact tests were used for comparing proportions and t tests for comparing means. We conducted subgroup analyses to assess the potential role of antiemetic use and the experience of morning sickness during the index pregnancy in the reporting of nausea and vomiting. We also conducted a per-protocol analysis that included in the Doxycycline arm only women who took at least one pill of doxycycline.
This study was approved by the Allendale Institutional Review Board, the Institute for Family Health Institutional Review Board, and the New York University School of Medicine Institutional Review Board.
3. Results Nine hundred and ten women were informed about the study, and 581 women enrolled including 278 in the Doxycycline arm and 303 in the No Doxycycline arm (Fig. 1). Thirty-seven percent and 30% of women in the respective groups did not enroll after being informed of the
Table 1 Demographic profile of women who were routinely prescribed a 7-day course of doxycycline and women who were not mean±S.D. or % (n/valid responses) Variable
Total n=581
Doxycycline n=278
No Doxycycline n=303
p-value
Age (years) Gestational age (days) Parity Had previous medical abortion Difficulty paying for abortion • Not hard at all • Somewhat hard • Very hard Race • White or Caucasian • Hispanic or Latina • Black or African-American • Multiracial • Other Education • Less than high school • High school graduate • College graduate • Advanced degree
27.1±6.2 46.6±8.9 2.6±1.9 31.7% (182/574)
27.1±6.3 47.9±7.9 2.6±2.1 34.1% (93/273)
27.1±6.1 45.3±9.6 2.6±1.8 29.6 (89/301)
.988 b .001 .941 .454 b .001
49.3% (262/531) 33.7% (179/531) 16.9% (90/531)
37.3% (100/268) 42.1% (113/268) 20.5% (55/268)
61.6% (162/263) 25.1% (66/263) 13.3% (35/263)
37.1% (213/574) 24.7% (142/574) 22.0% (126/574) 7.3% (42/574) 8.9% (51/574)
45.5% (125/275) 17.5% (48/275) 26.2% (72/275) 5.8% (16/275) 5.1% (14/275)
29.4% (88/299) 31.4% (94/299) 18.1% (54/299) 8.7% (26/299) 12.4% (37/299)
8.4% (48/569) 50.8% (289/569) 35.7% (203/569) 5.1% (29/569)
2.9% (8/273) 54.2% (148/273) 38.8% (106/273) 4.0% (11/273)
13.5% (40/296) 47.6% (141/296) 32.8% (97/296) 6.1% (18/296)
b .001
b .001
2-Tailed t test for continuous variables; Fisher’s exact for categorical data.
22
L.J. Frye et al. / Contraception 91 (2015) 19–24
Table 2 Adherence to a 7-day doxycycline regimen among women in the Doxycycline arm, % (n/valid responses) N= 278 Any pills taken Missed any doses • Missed N1 dose • Missed 2 consecutive doses More than 1 pill remaining at follow-up Extension of regimen (N8 days) Early termination of regimen (b 7 days) Perfect adherence a
97.5% (271/278) 44.2% (123/271) 29.1% (81/123) 9.7% (27/123) 35.3% (98/265) 9.3% (22/236) 34.3% (81/236) 28.3% (67/236)
a Perfect adherence is defined as self-report indicating no missed doses, 0 or 1 pill left at follow up, and pills taken for 7 or 8 days.
study, and an additional 2% and 4% of women respectively tried to enroll but were disqualified for reporting the use of unrelated antibiotics or refusal to consent. The study arms were similar with respect to mean age, parity, and experience of a previous medical abortion (Table 1). However mean gestational age, race, education, and difficulty paying for the abortion differed between the two arms. Among the 278 women in the Doxycycline arm seven (2.5%) reported not taking any of the doxycycline provided to them (Table 2). Forty-four percent reported having missed at least one dose and almost one-third missed more than one. Ten percent reported missing two consecutive doses and 34% reported stopping the regimen prior to 7 days. “Perfect adherence” was reported by 28.3% of women. The most common self-reported reasons for taking fewer pills than
prescribed among women who took no pills or had pills remaining at follow up (n=136) was that they were “still taking the pills” (52.9%), followed by “forgot to take the pills” (46.3%), and “had side effects” (12.5%). Compared to the No Doxycycline arm, there was a trend toward increased nausea in the Doxycycline arm (47.8% vs. 40.9%; p=.056) and a small but statistically significant difference in vomiting (25.2% vs 18.5%; p=.032) (Table 3). Additionally, both nausea and vomiting in the Doxycycline arm were significantly more likely to last longer than a single day (p=.04 and p=.002), and both were more likely to be rated as severe (p=.011 for nausea and p=.009 for vomiting). The No Doxycycline arm was more likely to report experiencing fever and chills (23.8% vs. 15.8% p=.011). One woman in the Doxycycline arm visited the Emergency Department due to an allergic reaction. She experienced a rash that covered 75% of her body and shortness of breath, and received Benadryl and prednisone. Based on the timing of the event and medical history it was presumed this allergic reaction was due to doxycycline. Seven other women experienced self-reported allergic reactions (4 Doxycycline, 3 No Doxycycline) which did not result in medical attention. When the 7 people in the Doxycycline arm who did not take any pills were excluded, the direction and significance level of the nausea and vomiting outcomes persisted [data not shown]. There was no difference in the two study arms with regard to taking prophylactic antiemetics (16.5% in the Doxycycline arm vs. 14.2% in the No Doxycycline arm, p=.251), and when limiting the sample to women who did not report
Table 3 Experience of side effects, severity, duration, and medications to treat side effects in the 7–14 days following medical abortion among the Doxycycline and No Doxycycline arms, % (n/valid responses)
Nausea • Severe • N 1 day Headache • Severe • N 1 day Vomiting • Severe • N 1 day Diarrhea • Severe • N 1 day Fever/chills • Severe • N 1 day Allergic reaction • Severe • N 1 day Yeast infection • Severe • N 1 day No side effects 2
One-tailed Fisher Exact test.
Total N=581
Doxycycline N=278
No Doxycycline N=303
p-value 2
44.2% (257/581) 6.7% (39/581) 24.8% (128/516) 28.2% (164/581) 3.5% (20/576) 18.2% (99/545) 21.7% (126/581) 4.8% (28/578) 5.6% (31/555) 19.1% 111/581 1.4% (8/581) 7.0% (39/559) 20.0% (116/581) 1.9% (11/581) 6.3% (35/557) 1.4% (8/581) 0.2% (1/581) 0.9% (5/579) 1.2% (7/581) 0.2% (1/581) 0.9% (5/580) 34.6% (201/581)
47.8% (133/278) 9.4% (26/278) 28.5% (70/246) 30.6% (85/278) 3.3% (9/276) 19.3% (50/259) 25.2% (70/278) 7.2% (20/278) 8.7% (23/265) 16.9% (47/278) 1.1% (3/278) 7.0% (19/271) 15.8% (44/278) 1.4% (4/278) 4.4% (12/270) 1.8% (5/278) 0.4% (1/278) 1.4% (4/278) 1.4% (4/278) 0.0% (0/278) 1.1% (3/277) 37.8% (105/278)
40.9% (124/303) 4.3% (13/303) 21.5% (58/270) 26.1% (79/303) 3.7% (11/300) 17.1% (49/286) 18.5% (56/303) 2.7% (8/300) 2.8% (8/290) 21.1% (64/303) 1.7% (5/303) 6.9% (20/288) 23.8% (72/303) 2.3% (7/303) 8.1% (23/287) 1.0% (3/303) 0.0% (0/303) 0.3% (1/301) 1.0% (3/303) 0.3% (1/303) 0.7% 2/303 31.7% (96/303)
.056 .011 .042 .133 .486 .293 .032 .009 .002 .118 .411 .553 .011 .323 .059 .316 .479 .163 .453 .522 .458 .073
L.J. Frye et al. / Contraception 91 (2015) 19–24
taking prophylactic antiemetics (n= 492), the nausea and vomiting findings persist (nausea p=.073, vomiting p=.050) [data not shown]. About two-thirds of women in both groups reported experiencing morning sickness during their pregnancy; when the sample is limited to women who did not experience morning sickness (n= 206) a significant relationship for the nausea and vomiting outcomes remains (nausea p=.043, vomiting p=.021)[data not shown]. The data revealed high retention of counseling messages with over 85% of women reporting that they were told the number of pills to take, the reason for taking the pills, the importance of taking the pills with food, and the importance of finishing the prescription (Table 4). Information on side effects was the most poorly recalled part of counseling.
4. Discussion This study showed that women who were prescribed a 7-day course of doxycycline following medical abortion achieved moderate levels of adherence. The pills were given directly to the women; it is likely that if prescriptions were given instead (common practice for some clinics) adherence would be lower. In fact, a study of doxycycline given for pelvic inflammatory disease found that 28% of women never filled the prescription [27]. The study also measured side effects associated with doxycycline use. Despite only moderate adherence to the regimen, there was still a small but significant difference in rates of vomiting, and severe nausea and severe vomiting. These relationships persisted through subgroup analyses. It is unclear why the No Doxycycline arm would experience higher rates of fever and chills, however the reported rates in both arms are below what is found in the literature (41% in Winikoff 2008; 42% in Middleton 2005) [32,33]. Although rates of allergic reactions were not statistically significant, one woman experienced a severe enough reaction to send her to the emergency department repeatedly. We report significant differences in some background characteristics by study arm but as a non-randomized trial, this is not unexpected. The clinics which did not routinely prescribe doxycycline and thus were eligible for the No Doxycycline arm tended to be independent providers whose patient population might differ from those attending highTable 4 Antibiotic counseling messages reported received by women in the Doxycycline arm, % (n/valid responses) What counseling were you given about the medicine to prevent infections (antibiotics/doxycycline)? • The number of pills to take each day/when to take the pills • The reason you should take the medicine • If the pills should be taken with food • The importance of taking all the pills • What side effects to expect
N=278 94.6% (262/277) 91.7% (254/277) 89.2% (247/277) 86.6% (240/277) 64.6% (179/277)
23
volume clinics. Additionally, we did not control practices but rather observed the standard of care, which differed slightly by site. Notably there were site-specific differences in the provision of antiemetics and analgesic regimens. Therefore we cannot rule out that other factors contributed to the differences in nausea and vomiting. This study relied on self-reported data. Because of poor recall, reporting of side effects may be different than what was actually experienced. Conversely, it is likely that the data we collected represents an over-estimation of the true adherence level as it is well documented that patients overreport their adherence to medical regimens [29,34,35]. The generalizability of the study findings is potentially limited by selection bias. While the study design attempted to compensate for the known low proportion of women who attend follow up after medical abortion by including a webbased mechanism for data collection, this was only used by 13% of the sample. It is possible that the side effect and adherence profiles of participants who did not enroll differed from those who provided data. While rates of side effects are consistent with the literature [31], it is possible that women who do not return for their follow-up visit are also less adherent to antibiotic regimens, so the adherence levels reported in the study may be conservative. Several months into recruitment there was a national shortage of doxycycline resulting from increased demand and manufacturing issues [36]. This resulted in some of the clinics in the Doxycycline arm switching to a different, less expensive antibiotic regimen, necessitating the halt of recruitment. This shortage has implications for the universal provision of doxycycline following medical abortion because it is no longer an inexpensive option. In fact, the rising cost of doxycycline provided PPFA an opportunity to review their antibiotic guidelines and develop new ones that were approved in the fall of 2013. These new guidelines include a single dose of doxycycline, azithromycin, or metronidazole, which may address some of the adherence issues found in this study. A sizable proportion of participants expressed confusion over what medications they had received at the clinic. Women swallow mifepristone at the clinic and are commonly given misoprostol and sometimes analgesics and antibiotics to take home as well as prescriptions for narcotics and anti-emetics. Given the limits of any patients’ ability to follow multiple and varied instructions, it is worth considering the impact of a 7-day course of doxycycline added to this cocktail when efficacy against severe infection is not robustly demonstrated in the literature. Similarly, shortened courses of prophylactic antibiotics after surgical abortion have been proven effective yet still are not universally adopted [37,38]. These data show that provision of a 7-day course of doxycycline following medical abortion is associated with increased vomiting. In the absence of robust efficacy evidence, we urge that the advantages and disadvantages of routine doxycycline provision be considered when providers review medical abortion protocols. If providers choose to continue doxycycline provision, they might select a shorter course.
24
L.J. Frye et al. / Contraception 91 (2015) 19–24
Acknowledgments Reprint requests should be made to [email protected]. The authors acknowledge the contributions of Philicia Castillo and Rayna Sobieski to central study coordination. We also thank the site study coordinators: Mary Alison Cooper, Rose Ali, Martha Olvera-Reinhofer, Michelle Arneson, Gloria Nesmith, Lauren Mitchell, Melissa Castro, Lia Cassanego, Lucia McLendon, Zanny Regas-Riewarts, Lianne Salcido, Amie Anderson, Bethany Elliott, MaryAnn Kozlowski, Alisa Claeys, Chan Chao, Staci Denis, and Mona Walia. Funding for this project was supplied by an anonymous donor without financial interests in the outcome of this study. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of Planned Parenthood Federation of America.
References [1] Fjerstad M, Sivin I, Lichtenberg ES, Trussell J, Cleland K, Cullins V. Effectiveness of medical abortion with mifepristone and buccal misoprostol through 59 gestational days. Contraception 2009;80(3):282–6. [2] Raymond EG, Shannon C, Weaver MA, Winikoff B. First-trimester medical abortion with mifepristone 200 mg and misoprostol: a systematic review. Contraception 2013;87(1):26–37. [3] Karki C, Pokharel H, Kushwaha A, Manandhar D, Bracken H, Winikoff B. Acceptability and feasibility of medical abortion in Nepal. Int J Gynaecol Obstet 2009;106(1):39–42. [4] Teal SB, Dempsey-Fanning A, Westhoff C. Predictors of acceptability of medication abortion. Contraception 2007;75(3):224–9. [5] Teal SB, Harken T, Sheeder J, Westhoff C. Efficacy, acceptability and safety of medication abortion in low-income, urban Latina women. Contraception 2009;80(5):479–83. [6] Winikoff B, Sivin I, Coyaji KJ, Cabezas E, Xiao B, Gu S, et al. Safety, efficacy, and acceptability of medical abortion in China, Cuba, and India: a comparative trial of mifepristone-misoprostol versus surgical abortion. Am J Obstet Gynecol 1997;176(2):431–7. [7] Woldetsadik MA, Sendekie TY, White MT, Zegeye DT. Client preferences and acceptability for medical abortion and MVA as early pregnancy termination method in northwest Ethiopia. Reprod Health 2011;8:19. [8] Henderson JT, Hwang AC, Harper CC, Stewart FH. Safety of mifepristone abortions in clinical use. Contraception 2005;72(3):175–8. [9] Sitruk-Ware R. Mifepristone and misoprostol sequential regimen side effects, complications and safety. Contraception 2006;74(1):48–55. [10] Shannon C, Brothers LP, Philip NM, Winikoff B. Infection after medical abortion: a review of the literature. Contraception 2004;70(3):183–90. [11] Fjerstad M, Trussell J, Sivin I, Lichtenberg ES, Cullins V. Rates of serious infection after changes in regimens for medical abortion. N Engl J Med 2009;361(2):145–51. [12] Fischer M, Bhatnagar J, Guarner J, Reagan S, Hacker JK, Van Meter SH, et al. Fatal toxic shock syndrome associated with Clostridium sordellii after medical abortion. N Engl J Med 2005;353(22):2352–60. [13] USFDA. Mifepristone US postmarketing adverse events summary through 04/30/2011; 2013. [14] Sinave C, Le Templier G, Blouin D, Léveillé F, Deland E. Toxic shock syndrome due to Clostridium sordellii: a dramatic postpartum and postabortion disease. Clin Infect Dis 2002;35(11):1441–3. [15] Trussell J, Nucatola D, Fjerstad M, Lichtenberg ES. Reduction in infection-related mortality since modifications in the regimen of medical abortion. Contraception 2014;89(3):193–6.
[16] Cleland K, Creinin MD, Nucatola D, Nshom M, Trussell J. Significant adverse events and outcomes after medical abortion. Obstet Gynecol 2013;121(1):166–71. [17] ACOG Practice Bulletin No. 143: Medical management of firsttrimester abortion. Obstet Gynecol 2014;123(3):676–92. [18] Clifford V, Daley A. Antibiotic prophylaxis in obstetric and gynaecological procedures: a review. Aust N Z J Obstet Gynaecol 2012;52(5):412–9. [19] Achilles SL, Reeves MF. Prevention of infection after induced abortion: release date October 2010: SFP guideline 20102. Contraception 2011;83 (4):295–309. [20] National Abortion Federation. 2014 Clinical Policy Guidelines. Washington, DC: NAF; 2014. [21] World Health Organization. Safe abortion: technical and policy guidance for health systems. 2nd ed. Geneva: WHO; 2012. [22] Ellertson C, Simonds W, Winikoff B, Springer K, Bagchi D. Providing mifepristone-misoprostol medical abortion: the view from the clinic. J Am Med Wom Assoc 1999;54(2):91–6. [23] ACOG Practice Bulletin No.67: Clinical management guidelines of obstetrician-gynecologists: medical management of abortion. Obstet Gynecol 2005;106(4):871–82. [24] Fjerstad M, Trussell J, Lichtenberg ES, Sivin I, Cullins V. Severity of infection following the introduction of new infection control measures for medical abortion. Contraception 2011;83(4):330–5. [25] Smith K, Leyden JJ. Safety of doxycycline and minocycline: a systematic review. Clin Ther 2005;27(9):1329–42. [26] Shapiro LE, Knowles SR, Shear NH. Comparative safety of tetracycline, minocycline, and doxycycline. Arch Dermatol 1997;133 (10):1224–30. [27] Brookoff D. Compliance with doxycycline therapy for outpatient treatment of pelvic inflammatory disease. South Med J 1994;87(11):1088–91. [28] Katz BP, Zwickl BW, Caine VA, Jones RB. Compliance with antibiotic therapy for Chlamydia trachomatis and Neisseria gonorrhoeae. Sex Transm Dis 1992;19(6):351–4. [29] Bachmann LH, Stephens J, Richey CM, Hook III EW. Measured versus self-reported compliance with doxycycline therapy for chlamydia-associated syndromes: high therapeutic success rates despite poor compliance. Sex Transm Dis 1999;26(5):272–8. [30] World Health Organization. Antimicrobial resistance: global report on surveillance. Geneva: WHO; 2014. [31] Winikoff B, Dzuba IG, Chong E, Goldberg AB, Lichtenberg ES, Ball C, et al. Extending outpatient medical abortion services through 70 days of gestational age. Obstet Gynecol 2012;120(5):1070–6. [32] Middleton T, Schaff E, Fielding SL, Scahill M, Shannon C, Westheimer E, et al. Randomized trial of mifepristone and buccal or vaginal misoprostol for abortion through 56 days of last menstrual period. Contraception 2005;72(5):328–32. [33] Winikoff B, Dzuba IG, Creinin MD, Crowden WA, Goldberg AB, Gonzales J, et al. Two distinct oral routes of misoprostol in mifepristone medical abortion: a randomized controlled trial. Obstet Gynecol 2008;112(6):1303–10. [34] Horne R, Weinman J, Barber N, Elliott R, Morgan M. Concordance, adherence and compliance in medicine taking. NCCSDO; 2005. [35] Waterhouse DM, Calzone KA, Mele C, Brenner DE. Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring. J Clin Oncol 1993;11(6):1189–97. [36] Sawyer SM, Belinda C, Carter Lucy S, Cooke Mary, Glazner Judith A, Massie John. Changing their minds with time: a comparison of hypothetical and actual reproductive behaviors in parents of children with cystic fibrosis. Pediatrics 2006;118(3):e649–56. [37] Lichtenberg ES, Shott S. A randomized clinical trial of prophylaxis for vacuum abortion: 3 versus 7 days of doxycycline. Obstet Gynecol 2003;101(4):726–31. [38] Levallois P, Rioux J-E. Prophylactic antibiotics for suction curettage abortion: Results of a clinical controlled trial. Am J Obstet Gynecol 1998;158(1):100–5.
Original research article
What happens when we routinely give doxycycline to medical abortion patients?☆,☆☆ Laura J. Frye⁎, Erica Chong, Beverly Winikoff, NCT01799252 Trial Investigators 1 Gynuity Health Projects, New York, NY 10010, USA Received 1 July 2014; revised 2 September 2014; accepted 2 September 2014
Abstract Objectives: Routine provision of antibiotics following medical abortion is common yet practitioners and professional societies differ on its utility. Our study compares the side effects experienced by women who were prescribed doxycycline following medical abortion to those who were not and assesses the adherence to one prescribed regimen. Study design: This was a prospective, observational, open-label study from a convenience sample. Women seeking medical abortion were enrolled in nine study sites, including four clinics that routinely prescribe a seven-day course of doxycycline (Doxycycline arm) and five clinics that do not routinely prescribe any antibiotics (No Doxycycline arm). Seven to fourteen days following the administration of mifepristone, women were asked to self-administer a computer-based survey. The survey asked about side effects experienced (both arms) and adherence to the regimen (Doxycycline arm only). Results: Five hundred eighty-one women were enrolled (278 in the Doxycycline arm and 303 in the No Doxycycline arm). There was a trend toward increased nausea in the Doxycycline arm (47.8% vs. 40.9%; p=.056) and a statistically significant difference in vomiting (25.2% vs. 18.5%; p=.032). Almost all women in the Doxycycline arm reported taking at least one pill, however only 28.3% reported “perfect adherence.” The most common reasons reported for taking fewer pills than instructed were that participants were still taking them (beyond 7 days) or that they forgot to take them. Conclusion: Women who were prescribed doxycycline following medical abortion reported moderate adherence and experienced significantly more vomiting than their counterparts. Implications: In the absence of robust evidence that prescribing 7 days of doxycycline following medical abortion is effective at reducing serious infections, these data can assist the public health community with deciding whether routine provision is the most appropriate strategy. © 2014 Elsevier Inc. All rights reserved. Keywords: Antibiotics; Doxycycline; Mifepristone; Misoprostol; Infection; Compliance
1. Introduction Early first trimester medical abortion is effective [1,2], highly acceptable to women [3–7], and safe [2,8,9]. The risk of infection following medical abortion is small, at less than 1% [10], with the rate of serious infection even lower, at .06/1000 [11]. Between 2001 and 2005, five women who had medical abortions in North America were infected with Clostridium ☆
The authors declare no conflicts of interest. Clinical Trial Registration Number: NCT01799252. ⁎ Corresponding author at: Gynuity Health Projects, 15 East 26th Street, Suite 801, New York, NY 10010. Tel.: +1 212 448 1230; fax: +1 212 448 1260. E-mail address: [email protected] (L.J. Frye). 1 Dr. Carol Ball, Dr. Lisa Harris, Dr. E. Steve Lichtenberg, Dr. Jules Marsh, Dr. Tamer Middleton, Dr. Amitasrigowri Murthy, Dr. Linda Prine. ☆☆
http://dx.doi.org/10.1016/j.contraception.2014.09.001 0010-7824/© 2014 Elsevier Inc. All rights reserved.
sordellii bacteria which caused toxic shock-like illness and death [12–14]. Although these fatal infections were extremely rare and no link was established between medical abortion and clostridia infection, Planned Parenthood Federation of America (PPFA) was prompted by these deaths to change its medical abortion protocols in 2006 to replace vaginal with buccal administration and to require routine antibiotic coverage [15]. These changes were widely adopted across the United States even among many independent providers. A retrospective review of data from PPFA showed a decrease over time in the serious infection rate after medical abortion following these protocol changes [11]. However, a subsequent report suggests that finding may have been a period effect [16,17]. Clifford and Daley estimated from the earlier PPFA data that 2500 women would need to be treated
20
L.J. Frye et al. / Contraception 91 (2015) 19–24
with antibiotics for a week to prevent one serious infection, and concluded that antibiotic prophylaxis for medical abortion is not warranted [18]. Indeed, recommendations from the World Health Organization, the Society of Family Planning, the American College of Obstetricians and Gynecologists, the National Abortion Federation, and the United States Food and Drug Administration all do not support routine use of preventive antibiotics for medical abortion [19–23]. To be clear, there is robust evidence of the efficacy of routine antibiotic prophylaxis following surgical abortion; however no clinical trials provide the same evidence base for medical abortion. Despite the lack of conclusive evidence on the efficacy of routine antibiotic treatment following medical abortion, this practice is common [24]. Considering that over 85,000 medical abortions are performed in the US each year, the advantages and disadvantages of any associated routine practice are amplified. At the time the study was conducted, doxycycline was the most widely used antibiotic following medical abortion and was typically given in a seven-day treatment dose of 100 mg twice daily. This treatment does not come without costs financially, logistically, and in terms of side effects. Doxycycline is associated with gastrointestinal, skin, and urogenital side effects [25] and while allergy is rare, it can include hypersensitivity syndrome reaction and serum sickness-like reaction [26]. Difficulty adhering to any antibiotic regimen would have implications for the benefits of routine prescription. Extensive evidence suggests that patient compliance with antibiotics is low [27,28]. A study of patients treated with doxycycline for Chlamydia found that 16% of patients fully adhered (as measured by electronic pill bottle cap) and 67% of patients had at least one instance of 24 h between doses [29]. While beyond the scope of this paper, public health concerns around the overuse of antibiotics and the development of antimicrobial resistance are important when examining the pros and cons of routine provision [30]. In summary, robust evidence is lacking on the effectiveness of doxycycline treatment following medical abortion but most women in the US receive it. This study seeks to add to the literature evidence of the side effects women experience with post-abortion doxycycline treatment and their adherence to the regimen.
2. Materials and methods This was a non-randomized, observational study conducted from October 2012 to December 2013. Participants were recruited from 9 clinics in the United States. Four clinics routinely prescribed doxycycline following medical abortion and five did not. All sites scheduled follow up visits at seven days following administration of mifepristone and, in the doxycycline arm, all sites dispensed the doxycycline directly to participants who were counseled to start the doxycycline the day they took mifepristone. Sites prescribed
ancillary medicines (analgesics, antiemetics, etc.) according to their regular protocols or patient needs. Women presenting for a medical abortion in the study clinics who could read English or Spanish were eligible for the study. Any woman who was currently taking antibiotics for reasons unrelated to her medical abortion was excluded as was any woman who had previously participated in this study. On the day they took their mifepristone in the clinic, eligible women were provided with a study information card which listed a unique Study ID number and a website where they could access a computer-based questionnaire 7–14 days later. Women could take the questionnaire either at the clinic during their follow-up visits, or off-site wherever they had an internet connection. If participants chose to take the questionnaire at the follow-up visit, they were given a laptop to self-administer it in a private location. Women who did not return for follow-up were called and reminded of the possibility of participating in the study using the website and Study ID provided to them. A woman was considered to be enrolled in the study once she accessed the website and gave her informed consent. The questionnaire used DatStat Illume v. 4.11 software and took 5–10 minutes to complete. The survey contained questions on demographics, medications dispensed or prescribed, side effects experienced and adherence to the doxycycline regimen (Doxycycline arm only). All questions related to the time period between taking the mifepristone in the clinic and the date of survey administration (7–14 days later). Women were provided with a list of side effects including: nausea, headache, vomiting, diarrhea, fever/chills, allergic reaction (rash, itching, wheezing) and yeast infection. They were asked to check each side effect experienced and had the option of writing in any others. We did not consider pain or bleeding to be side effects as they are the main effects of treatment. For each side effect women reported they were asked the severity, duration, and whether they took medications to treat it. In the Doxycycline arm, adherence was measured by asking the dates when women started and stopped taking doxycycline, if any doses were missed, if two consecutive doses were missed, and the number of pills remaining at the time of survey completion. All women who took fewer pills than prescribed were asked to provide a reason. “Perfect adherence” was defined as not missing any doses, having 0 or 1 pill remaining at the follow-up visit, and taking the medicine for 7 or 8 days. Women were also asked about elements of antibiotic counseling they received. No information on infection rates was obtained as the goals of the study were to focus on adherence and side effects of a commonly used antibiotic regimen. A sample size of 610 women (305 in each group) would be sufficient for a one-sided test to detect with 80% power a 10% lower nausea rate in the No Doxycycline arm (assuming a 50% nausea rate in the Doxycycline arm) [31]. A sample size of 305 in the Doxycycline arm would also allow an estimation of
L.J. Frye et al. / Contraception 91 (2015) 19–24
21
Fig. 1. Enrollment flow diagram.
adherence proportions with precisions of ± 6 percentage points. During the trial period there were two national doxycycline shortages which halted recruitment in the Doxycycline arm and limited the sample size. Statistical Package for Social Scientists (SPSS) v 19 and STATA v 12 were used for statistical analyses. Fisher's exact tests were used for comparing proportions and t tests for comparing means. We conducted subgroup analyses to assess the potential role of antiemetic use and the experience of morning sickness during the index pregnancy in the reporting of nausea and vomiting. We also conducted a per-protocol analysis that included in the Doxycycline arm only women who took at least one pill of doxycycline.
This study was approved by the Allendale Institutional Review Board, the Institute for Family Health Institutional Review Board, and the New York University School of Medicine Institutional Review Board.
3. Results Nine hundred and ten women were informed about the study, and 581 women enrolled including 278 in the Doxycycline arm and 303 in the No Doxycycline arm (Fig. 1). Thirty-seven percent and 30% of women in the respective groups did not enroll after being informed of the
Table 1 Demographic profile of women who were routinely prescribed a 7-day course of doxycycline and women who were not mean±S.D. or % (n/valid responses) Variable
Total n=581
Doxycycline n=278
No Doxycycline n=303
p-value
Age (years) Gestational age (days) Parity Had previous medical abortion Difficulty paying for abortion • Not hard at all • Somewhat hard • Very hard Race • White or Caucasian • Hispanic or Latina • Black or African-American • Multiracial • Other Education • Less than high school • High school graduate • College graduate • Advanced degree
27.1±6.2 46.6±8.9 2.6±1.9 31.7% (182/574)
27.1±6.3 47.9±7.9 2.6±2.1 34.1% (93/273)
27.1±6.1 45.3±9.6 2.6±1.8 29.6 (89/301)
.988 b .001 .941 .454 b .001
49.3% (262/531) 33.7% (179/531) 16.9% (90/531)
37.3% (100/268) 42.1% (113/268) 20.5% (55/268)
61.6% (162/263) 25.1% (66/263) 13.3% (35/263)
37.1% (213/574) 24.7% (142/574) 22.0% (126/574) 7.3% (42/574) 8.9% (51/574)
45.5% (125/275) 17.5% (48/275) 26.2% (72/275) 5.8% (16/275) 5.1% (14/275)
29.4% (88/299) 31.4% (94/299) 18.1% (54/299) 8.7% (26/299) 12.4% (37/299)
8.4% (48/569) 50.8% (289/569) 35.7% (203/569) 5.1% (29/569)
2.9% (8/273) 54.2% (148/273) 38.8% (106/273) 4.0% (11/273)
13.5% (40/296) 47.6% (141/296) 32.8% (97/296) 6.1% (18/296)
b .001
b .001
2-Tailed t test for continuous variables; Fisher’s exact for categorical data.
22
L.J. Frye et al. / Contraception 91 (2015) 19–24
Table 2 Adherence to a 7-day doxycycline regimen among women in the Doxycycline arm, % (n/valid responses) N= 278 Any pills taken Missed any doses • Missed N1 dose • Missed 2 consecutive doses More than 1 pill remaining at follow-up Extension of regimen (N8 days) Early termination of regimen (b 7 days) Perfect adherence a
97.5% (271/278) 44.2% (123/271) 29.1% (81/123) 9.7% (27/123) 35.3% (98/265) 9.3% (22/236) 34.3% (81/236) 28.3% (67/236)
a Perfect adherence is defined as self-report indicating no missed doses, 0 or 1 pill left at follow up, and pills taken for 7 or 8 days.
study, and an additional 2% and 4% of women respectively tried to enroll but were disqualified for reporting the use of unrelated antibiotics or refusal to consent. The study arms were similar with respect to mean age, parity, and experience of a previous medical abortion (Table 1). However mean gestational age, race, education, and difficulty paying for the abortion differed between the two arms. Among the 278 women in the Doxycycline arm seven (2.5%) reported not taking any of the doxycycline provided to them (Table 2). Forty-four percent reported having missed at least one dose and almost one-third missed more than one. Ten percent reported missing two consecutive doses and 34% reported stopping the regimen prior to 7 days. “Perfect adherence” was reported by 28.3% of women. The most common self-reported reasons for taking fewer pills than
prescribed among women who took no pills or had pills remaining at follow up (n=136) was that they were “still taking the pills” (52.9%), followed by “forgot to take the pills” (46.3%), and “had side effects” (12.5%). Compared to the No Doxycycline arm, there was a trend toward increased nausea in the Doxycycline arm (47.8% vs. 40.9%; p=.056) and a small but statistically significant difference in vomiting (25.2% vs 18.5%; p=.032) (Table 3). Additionally, both nausea and vomiting in the Doxycycline arm were significantly more likely to last longer than a single day (p=.04 and p=.002), and both were more likely to be rated as severe (p=.011 for nausea and p=.009 for vomiting). The No Doxycycline arm was more likely to report experiencing fever and chills (23.8% vs. 15.8% p=.011). One woman in the Doxycycline arm visited the Emergency Department due to an allergic reaction. She experienced a rash that covered 75% of her body and shortness of breath, and received Benadryl and prednisone. Based on the timing of the event and medical history it was presumed this allergic reaction was due to doxycycline. Seven other women experienced self-reported allergic reactions (4 Doxycycline, 3 No Doxycycline) which did not result in medical attention. When the 7 people in the Doxycycline arm who did not take any pills were excluded, the direction and significance level of the nausea and vomiting outcomes persisted [data not shown]. There was no difference in the two study arms with regard to taking prophylactic antiemetics (16.5% in the Doxycycline arm vs. 14.2% in the No Doxycycline arm, p=.251), and when limiting the sample to women who did not report
Table 3 Experience of side effects, severity, duration, and medications to treat side effects in the 7–14 days following medical abortion among the Doxycycline and No Doxycycline arms, % (n/valid responses)
Nausea • Severe • N 1 day Headache • Severe • N 1 day Vomiting • Severe • N 1 day Diarrhea • Severe • N 1 day Fever/chills • Severe • N 1 day Allergic reaction • Severe • N 1 day Yeast infection • Severe • N 1 day No side effects 2
One-tailed Fisher Exact test.
Total N=581
Doxycycline N=278
No Doxycycline N=303
p-value 2
44.2% (257/581) 6.7% (39/581) 24.8% (128/516) 28.2% (164/581) 3.5% (20/576) 18.2% (99/545) 21.7% (126/581) 4.8% (28/578) 5.6% (31/555) 19.1% 111/581 1.4% (8/581) 7.0% (39/559) 20.0% (116/581) 1.9% (11/581) 6.3% (35/557) 1.4% (8/581) 0.2% (1/581) 0.9% (5/579) 1.2% (7/581) 0.2% (1/581) 0.9% (5/580) 34.6% (201/581)
47.8% (133/278) 9.4% (26/278) 28.5% (70/246) 30.6% (85/278) 3.3% (9/276) 19.3% (50/259) 25.2% (70/278) 7.2% (20/278) 8.7% (23/265) 16.9% (47/278) 1.1% (3/278) 7.0% (19/271) 15.8% (44/278) 1.4% (4/278) 4.4% (12/270) 1.8% (5/278) 0.4% (1/278) 1.4% (4/278) 1.4% (4/278) 0.0% (0/278) 1.1% (3/277) 37.8% (105/278)
40.9% (124/303) 4.3% (13/303) 21.5% (58/270) 26.1% (79/303) 3.7% (11/300) 17.1% (49/286) 18.5% (56/303) 2.7% (8/300) 2.8% (8/290) 21.1% (64/303) 1.7% (5/303) 6.9% (20/288) 23.8% (72/303) 2.3% (7/303) 8.1% (23/287) 1.0% (3/303) 0.0% (0/303) 0.3% (1/301) 1.0% (3/303) 0.3% (1/303) 0.7% 2/303 31.7% (96/303)
.056 .011 .042 .133 .486 .293 .032 .009 .002 .118 .411 .553 .011 .323 .059 .316 .479 .163 .453 .522 .458 .073
L.J. Frye et al. / Contraception 91 (2015) 19–24
taking prophylactic antiemetics (n= 492), the nausea and vomiting findings persist (nausea p=.073, vomiting p=.050) [data not shown]. About two-thirds of women in both groups reported experiencing morning sickness during their pregnancy; when the sample is limited to women who did not experience morning sickness (n= 206) a significant relationship for the nausea and vomiting outcomes remains (nausea p=.043, vomiting p=.021)[data not shown]. The data revealed high retention of counseling messages with over 85% of women reporting that they were told the number of pills to take, the reason for taking the pills, the importance of taking the pills with food, and the importance of finishing the prescription (Table 4). Information on side effects was the most poorly recalled part of counseling.
4. Discussion This study showed that women who were prescribed a 7-day course of doxycycline following medical abortion achieved moderate levels of adherence. The pills were given directly to the women; it is likely that if prescriptions were given instead (common practice for some clinics) adherence would be lower. In fact, a study of doxycycline given for pelvic inflammatory disease found that 28% of women never filled the prescription [27]. The study also measured side effects associated with doxycycline use. Despite only moderate adherence to the regimen, there was still a small but significant difference in rates of vomiting, and severe nausea and severe vomiting. These relationships persisted through subgroup analyses. It is unclear why the No Doxycycline arm would experience higher rates of fever and chills, however the reported rates in both arms are below what is found in the literature (41% in Winikoff 2008; 42% in Middleton 2005) [32,33]. Although rates of allergic reactions were not statistically significant, one woman experienced a severe enough reaction to send her to the emergency department repeatedly. We report significant differences in some background characteristics by study arm but as a non-randomized trial, this is not unexpected. The clinics which did not routinely prescribe doxycycline and thus were eligible for the No Doxycycline arm tended to be independent providers whose patient population might differ from those attending highTable 4 Antibiotic counseling messages reported received by women in the Doxycycline arm, % (n/valid responses) What counseling were you given about the medicine to prevent infections (antibiotics/doxycycline)? • The number of pills to take each day/when to take the pills • The reason you should take the medicine • If the pills should be taken with food • The importance of taking all the pills • What side effects to expect
N=278 94.6% (262/277) 91.7% (254/277) 89.2% (247/277) 86.6% (240/277) 64.6% (179/277)
23
volume clinics. Additionally, we did not control practices but rather observed the standard of care, which differed slightly by site. Notably there were site-specific differences in the provision of antiemetics and analgesic regimens. Therefore we cannot rule out that other factors contributed to the differences in nausea and vomiting. This study relied on self-reported data. Because of poor recall, reporting of side effects may be different than what was actually experienced. Conversely, it is likely that the data we collected represents an over-estimation of the true adherence level as it is well documented that patients overreport their adherence to medical regimens [29,34,35]. The generalizability of the study findings is potentially limited by selection bias. While the study design attempted to compensate for the known low proportion of women who attend follow up after medical abortion by including a webbased mechanism for data collection, this was only used by 13% of the sample. It is possible that the side effect and adherence profiles of participants who did not enroll differed from those who provided data. While rates of side effects are consistent with the literature [31], it is possible that women who do not return for their follow-up visit are also less adherent to antibiotic regimens, so the adherence levels reported in the study may be conservative. Several months into recruitment there was a national shortage of doxycycline resulting from increased demand and manufacturing issues [36]. This resulted in some of the clinics in the Doxycycline arm switching to a different, less expensive antibiotic regimen, necessitating the halt of recruitment. This shortage has implications for the universal provision of doxycycline following medical abortion because it is no longer an inexpensive option. In fact, the rising cost of doxycycline provided PPFA an opportunity to review their antibiotic guidelines and develop new ones that were approved in the fall of 2013. These new guidelines include a single dose of doxycycline, azithromycin, or metronidazole, which may address some of the adherence issues found in this study. A sizable proportion of participants expressed confusion over what medications they had received at the clinic. Women swallow mifepristone at the clinic and are commonly given misoprostol and sometimes analgesics and antibiotics to take home as well as prescriptions for narcotics and anti-emetics. Given the limits of any patients’ ability to follow multiple and varied instructions, it is worth considering the impact of a 7-day course of doxycycline added to this cocktail when efficacy against severe infection is not robustly demonstrated in the literature. Similarly, shortened courses of prophylactic antibiotics after surgical abortion have been proven effective yet still are not universally adopted [37,38]. These data show that provision of a 7-day course of doxycycline following medical abortion is associated with increased vomiting. In the absence of robust efficacy evidence, we urge that the advantages and disadvantages of routine doxycycline provision be considered when providers review medical abortion protocols. If providers choose to continue doxycycline provision, they might select a shorter course.
24
L.J. Frye et al. / Contraception 91 (2015) 19–24
Acknowledgments Reprint requests should be made to [email protected]. The authors acknowledge the contributions of Philicia Castillo and Rayna Sobieski to central study coordination. We also thank the site study coordinators: Mary Alison Cooper, Rose Ali, Martha Olvera-Reinhofer, Michelle Arneson, Gloria Nesmith, Lauren Mitchell, Melissa Castro, Lia Cassanego, Lucia McLendon, Zanny Regas-Riewarts, Lianne Salcido, Amie Anderson, Bethany Elliott, MaryAnn Kozlowski, Alisa Claeys, Chan Chao, Staci Denis, and Mona Walia. Funding for this project was supplied by an anonymous donor without financial interests in the outcome of this study. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of Planned Parenthood Federation of America.
References [1] Fjerstad M, Sivin I, Lichtenberg ES, Trussell J, Cleland K, Cullins V. Effectiveness of medical abortion with mifepristone and buccal misoprostol through 59 gestational days. Contraception 2009;80(3):282–6. [2] Raymond EG, Shannon C, Weaver MA, Winikoff B. First-trimester medical abortion with mifepristone 200 mg and misoprostol: a systematic review. Contraception 2013;87(1):26–37. [3] Karki C, Pokharel H, Kushwaha A, Manandhar D, Bracken H, Winikoff B. Acceptability and feasibility of medical abortion in Nepal. Int J Gynaecol Obstet 2009;106(1):39–42. [4] Teal SB, Dempsey-Fanning A, Westhoff C. Predictors of acceptability of medication abortion. Contraception 2007;75(3):224–9. [5] Teal SB, Harken T, Sheeder J, Westhoff C. Efficacy, acceptability and safety of medication abortion in low-income, urban Latina women. Contraception 2009;80(5):479–83. [6] Winikoff B, Sivin I, Coyaji KJ, Cabezas E, Xiao B, Gu S, et al. Safety, efficacy, and acceptability of medical abortion in China, Cuba, and India: a comparative trial of mifepristone-misoprostol versus surgical abortion. Am J Obstet Gynecol 1997;176(2):431–7. [7] Woldetsadik MA, Sendekie TY, White MT, Zegeye DT. Client preferences and acceptability for medical abortion and MVA as early pregnancy termination method in northwest Ethiopia. Reprod Health 2011;8:19. [8] Henderson JT, Hwang AC, Harper CC, Stewart FH. Safety of mifepristone abortions in clinical use. Contraception 2005;72(3):175–8. [9] Sitruk-Ware R. Mifepristone and misoprostol sequential regimen side effects, complications and safety. Contraception 2006;74(1):48–55. [10] Shannon C, Brothers LP, Philip NM, Winikoff B. Infection after medical abortion: a review of the literature. Contraception 2004;70(3):183–90. [11] Fjerstad M, Trussell J, Sivin I, Lichtenberg ES, Cullins V. Rates of serious infection after changes in regimens for medical abortion. N Engl J Med 2009;361(2):145–51. [12] Fischer M, Bhatnagar J, Guarner J, Reagan S, Hacker JK, Van Meter SH, et al. Fatal toxic shock syndrome associated with Clostridium sordellii after medical abortion. N Engl J Med 2005;353(22):2352–60. [13] USFDA. Mifepristone US postmarketing adverse events summary through 04/30/2011; 2013. [14] Sinave C, Le Templier G, Blouin D, Léveillé F, Deland E. Toxic shock syndrome due to Clostridium sordellii: a dramatic postpartum and postabortion disease. Clin Infect Dis 2002;35(11):1441–3. [15] Trussell J, Nucatola D, Fjerstad M, Lichtenberg ES. Reduction in infection-related mortality since modifications in the regimen of medical abortion. Contraception 2014;89(3):193–6.
[16] Cleland K, Creinin MD, Nucatola D, Nshom M, Trussell J. Significant adverse events and outcomes after medical abortion. Obstet Gynecol 2013;121(1):166–71. [17] ACOG Practice Bulletin No. 143: Medical management of firsttrimester abortion. Obstet Gynecol 2014;123(3):676–92. [18] Clifford V, Daley A. Antibiotic prophylaxis in obstetric and gynaecological procedures: a review. Aust N Z J Obstet Gynaecol 2012;52(5):412–9. [19] Achilles SL, Reeves MF. Prevention of infection after induced abortion: release date October 2010: SFP guideline 20102. Contraception 2011;83 (4):295–309. [20] National Abortion Federation. 2014 Clinical Policy Guidelines. Washington, DC: NAF; 2014. [21] World Health Organization. Safe abortion: technical and policy guidance for health systems. 2nd ed. Geneva: WHO; 2012. [22] Ellertson C, Simonds W, Winikoff B, Springer K, Bagchi D. Providing mifepristone-misoprostol medical abortion: the view from the clinic. J Am Med Wom Assoc 1999;54(2):91–6. [23] ACOG Practice Bulletin No.67: Clinical management guidelines of obstetrician-gynecologists: medical management of abortion. Obstet Gynecol 2005;106(4):871–82. [24] Fjerstad M, Trussell J, Lichtenberg ES, Sivin I, Cullins V. Severity of infection following the introduction of new infection control measures for medical abortion. Contraception 2011;83(4):330–5. [25] Smith K, Leyden JJ. Safety of doxycycline and minocycline: a systematic review. Clin Ther 2005;27(9):1329–42. [26] Shapiro LE, Knowles SR, Shear NH. Comparative safety of tetracycline, minocycline, and doxycycline. Arch Dermatol 1997;133 (10):1224–30. [27] Brookoff D. Compliance with doxycycline therapy for outpatient treatment of pelvic inflammatory disease. South Med J 1994;87(11):1088–91. [28] Katz BP, Zwickl BW, Caine VA, Jones RB. Compliance with antibiotic therapy for Chlamydia trachomatis and Neisseria gonorrhoeae. Sex Transm Dis 1992;19(6):351–4. [29] Bachmann LH, Stephens J, Richey CM, Hook III EW. Measured versus self-reported compliance with doxycycline therapy for chlamydia-associated syndromes: high therapeutic success rates despite poor compliance. Sex Transm Dis 1999;26(5):272–8. [30] World Health Organization. Antimicrobial resistance: global report on surveillance. Geneva: WHO; 2014. [31] Winikoff B, Dzuba IG, Chong E, Goldberg AB, Lichtenberg ES, Ball C, et al. Extending outpatient medical abortion services through 70 days of gestational age. Obstet Gynecol 2012;120(5):1070–6. [32] Middleton T, Schaff E, Fielding SL, Scahill M, Shannon C, Westheimer E, et al. Randomized trial of mifepristone and buccal or vaginal misoprostol for abortion through 56 days of last menstrual period. Contraception 2005;72(5):328–32. [33] Winikoff B, Dzuba IG, Creinin MD, Crowden WA, Goldberg AB, Gonzales J, et al. Two distinct oral routes of misoprostol in mifepristone medical abortion: a randomized controlled trial. Obstet Gynecol 2008;112(6):1303–10. [34] Horne R, Weinman J, Barber N, Elliott R, Morgan M. Concordance, adherence and compliance in medicine taking. NCCSDO; 2005. [35] Waterhouse DM, Calzone KA, Mele C, Brenner DE. Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring. J Clin Oncol 1993;11(6):1189–97. [36] Sawyer SM, Belinda C, Carter Lucy S, Cooke Mary, Glazner Judith A, Massie John. Changing their minds with time: a comparison of hypothetical and actual reproductive behaviors in parents of children with cystic fibrosis. Pediatrics 2006;118(3):e649–56. [37] Lichtenberg ES, Shott S. A randomized clinical trial of prophylaxis for vacuum abortion: 3 versus 7 days of doxycycline. Obstet Gynecol 2003;101(4):726–31. [38] Levallois P, Rioux J-E. Prophylactic antibiotics for suction curettage abortion: Results of a clinical controlled trial. Am J Obstet Gynecol 1998;158(1):100–5.
