Combination hormonal contraception is undergoing a transition away from the 21/7 regimen (21 days of combination estrogen and progestin therapy followed by 7 days of a hormone-free interval). Numerous studies have confirmed drawbacks to the institution of a 7-day hormone-free interval with today’s low-dose combination hormonal contraception regimens, including increased symptomatology (mood changes, headaches, and pelvic pain)1–3 and lack of pituitary–ovarian suppression resulting in follicular development, endogenous estradiol production, ovarian cyst formation, and possible ovulation.4–9
Studies have evaluated extended oral contraceptive (OC) regimens in which a hormone-free interval is instituted periodically and continuous regimens in which the hormone-free interval is eliminated.10–16 Clinicians have been using continuous OC regimens in an off-label fashion.17 Research has documented a reduction in hormone withdrawal symptoms with continuous OC regimens that eliminate the 7-day hormone-free interval.18–20 The most common side effect of extended/continuous OC regimens is breakthrough bleeding/spotting. Two studies have evaluated the effectiveness of instituting an abbreviated hormone-free interval of 3 to 4 days to manage the nuisance bleeding followed by resumption of active pills.10,11
Extended combination hormonal contraception regimens using the contraceptive vaginal ring also have been studied.21,22 In both of these extended-use studies, the ring was replaced every 3 weeks, with a 7-day hormone-free interval instituted after various intervals of extended use. NuvaRing (Organon, Inc., Roseland, NJ) is a 21/7 contraceptive vaginal ring containing ethinyl estradiol and the progestin etonogestrel. The manufacturer of the ring has data to confirm contraceptive efficacy with up to 28-day use (NuvaRing Physician Insert), and circulating levels for up to 35 days of extended use have been reported.23 Use beyond 4 weeks is not recommended by the manufacturer (NuvaRing Physician Insert). The objectives of this study were to compare symptomatology and bleeding patterns of standard 21/7 ring use with continuous ring use replaced monthly and to test whether breakthrough bleeding/spotting associated with continuous ring use can be reduced with the institution of an abbreviated hormone-free interval.
Women age 18–45 who had been using 21/7 combination hormonal contraception (OCs, the transvaginal ring, or the transdermal contraceptive patch) for a minimum of 2 months were recruited from the community from those responding to advertisements into this study, which used a continuous regimen of the transvaginal contraceptive ring NuvaRing (Organon, Inc.). Women were excluded if they had a body mass index of 38 kg/m2 or greater, if there were contraindications to using combination hormonal contraception, if they had tobacco use of more than 10 cigarettes per day, if they were taking antiretroviral therapy (because of many drug interactions), if they were using other estrogen-containing products or herbal products that contained phytoestrogens, or if they desired pregnancy within the subsequent year.
Each participant was required to monitor daily symptoms, ring usage, and breakthrough bleeding/spotting. Participants kept the Scott & White Daily Diary of Symptoms, which scored occurrence and degree of menstrual flow, pelvic pain, headaches, and mood. Participants scored their flow on a scale of 0 to 4: zero for no spotting or bleeding; 1 for minimal spotting, not requiring sanitary protection; 2 for mild bleeding with an estimate of no more than three pads/tampons soaked per 24 hours; 3 for moderate bleeding with an estimate of more than three but fewer than six pads/tampons soaked; and 4 for heavy flow with an estimate of more than six pads/tampons soaked. This scale has been used in previous studies to quantify flow subjectively rather than simply recording spotting or bleeding.11,20 However, this definition is in keeping with the use of spotting and up to three levels of bleeding proposed by Gerlinger et al.24 Pelvic pain, headaches, and mood were scored on a visual analog scale of 0–10. The mood score represented a composite of the common symptoms of depression, anxiety, and irritability.25
Women had entry physical and pelvic examinations as well as a vaginal ultrasound examination to document normal pelvic anatomy. Participants with ovarian cysts greater than 2.5 cm or an endometrial thickness greater than 8 mm were excluded. Although there are no data stating that these specific measurements are abnormal findings in patients on combination hormonal contraception, they were chosen to rule out significant ovarian and uterine pathology that could cause breakthrough bleeding/spotting. Pap tests were performed based on American College of Obstetricians and Gynecologists guidelines. The study was approved by the Scott and White institutional review board, and written informed consent was obtained.
Participants who were using OCs or the transdermal contraceptive patch completed two standard 21/7 baseline cycles of the transvaginal contraceptive ring, and those who were using the contraceptive ring at enrollment completed one 21/7 baseline cycle. This preextension phase allowed assessment of acceptability of the device in those unaccustomed to its use and baseline data for all participants. This baseline phase also screened for the appropriateness of candidates for the study. Only those women who desired to continue transvaginal contraceptive ring use and who could correctly complete all necessary research instruments/diaries were allowed to enter the continuous use phase.
Those who met the above criteria were allowed to enter into the continuous use phase of the trial. Participants were asked to complete 6 months of continuous transvaginal ring therapy, replacing the ring monthly. It was anticipated that the majority of participants would have breakthrough bleeding/spotting episodes at some time during the continuous use phase. To assess management of breakthrough bleeding/spotting, participants were randomly divided into two groups using a sealed envelope assignment system with four blocks of 20 participants, with 10 assigned to each group using a computerized random number sequence blinded to study staff until each assignment was opened.
Women in group 1 continued to use the monthly transvaginal contraceptive ring device for 6 months without any hormone-free days. For example, a patient inserting the first continuous ring on October 17 would replace the device on November 17, December 17, January 17, February 17, and March 17. Even if breakthrough bleeding/spotting developed, participants were instructed not to alter their monthly replacement. Women in group 2 used continuous monthly rings for 6 months as described for group 1, but if a breakthrough bleeding/spotting episode of more than 5 days occurred, the participant was allowed to remove the contraceptive ring, store it in the original sachet for 4 days, and then reinsert the stored ring. At the completion of the current ring, the device was removed and discarded and a new ring was inserted on the same calendar day of the month as the current ring, keeping the original start day of the month. For example, if a patient inserted the ring on the 19th day of each month, for example on November 19th, but developed breakthrough bleeding/spotting from November 30–December 4, she would remove the ring on December 5, leave it out and store it on December 6, 7, and 8, and reinsert that ring on December 9. A new ring would be inserted on December 19. Using this method, the patient always knew that she was to exchange her ring on the same calendar day of each month.
Participant characteristics were expressed as means with standard deviations, and comparisons of randomized groups used analysis of variance (ANOVA) with Wilks’ test or Pearson’s χ2 test using Statistica software (StatSoft, Tulsa, OK). The planned analyses focused on a comparison of days and average daily flow scores within participants for a 21/7-day pattern of use during the last cycle of the baseline phase and the average for 6 months of subsequent continuous use (using Wilcoxon matched pairs test), a comparison between groups of percentage of flow-free days and average daily flow scores during the 6 monthly intervals (using ANOVA with repeated measures), and a comparison between groups of the bleeding episodes meeting criteria for intervention (ANOVA with repeated measures). In addition, a comparison was planned regarding preference on management of breakthrough bleeding/spotting and comparisons of hormone withdrawal symptoms, including pelvic pain, headaches, and mood score, within participants during 21/7 and subsequent continuous use.
Based on a previous publication11 with an extended OC regimen of 24 weeks compared with the standard 21/7, we expected to find that about 20% (mean±standard deviation 20.1±10.0% for a sample of 102 participants) of the days of a 21/7 cycle would have spotting or bleeding, compared with 13% (13.0±11.6%) for these participants subsequently using an extended regimen. A sample size of 35 participants would have a power of 0.8 to detect such a difference at P<.05 within participants for each group of continuous ring users. Before this study, we had no data to assess comparisons of continuous transvaginal delivery of contraceptive hormones for 6 months to a pattern that incorporates a 4-day hormone-free interval as a treatment for breakthrough bleeding/spotting. For data sets of percentages and average flow scores with heterogeneity of variance using Levene’s test (Statistica [StatSoft]), arcsine transformation was used for analysis, with presentation of results as geometric means with 95% confidence intervals.
Enrollment began in September 2006, and 6 months of continuous use was completed for all participants by December 2007. Ninety-two participants signed consents. Two had a body mass index greater than 38 kg/m2, and four moved away or declined further participation for personal reasons, leaving 86 women enrolled in the baseline phase. Eleven participants failed to complete this baseline phase. Six participants had ovarian cysts 2.5 cm or greater, two participants had endometrial thickness greater than 8 mm, two participants decided that they could not make the study visits, and one participant withdrew because of mood lability. The remaining 75 participants completed the 21/7 baseline phase of the study and entered the continuous phase of ring use. Current contraceptive use included 38 on OCs, 29 using the vaginal ring, and eight transdermal patch uses. Of the 75 participants, 38 were randomized to group 1 and 37 to group 2. One participant in group 1 withdrew from the continuous phase because of vaginal discharge without providing diaries or other information. Of the 74 participants in the continuous phase contributing results to this trial, 37 were randomized to group 1 (to receive no intervention for breakthrough bleeding/spotting) and 37 were randomized to group 2 (to take a 4-day hormone-free interval if breakthrough bleeding/spotting exceeded 5 days). Data analysis is based on these 74 patients. These two groups of participants had similar demographic, age, body size, and menstrual flow characteristics during the 21/7 baseline phase of transvaginal ring use (Table 1).
Eight (22%) of the 37 participants assigned to group 1 withdrew from the study before completing the 6 months of continuous use: four because of side effects including mood lability, headache, and/or bleeding, two who were unable to attend scheduled visits, one who desired pregnancy, and one who was lost to follow-up. One (3%) of the 37 participants randomized to group 2 withdrew because of a desire to become pregnant. The difference in study completion for the two groups was significant (P=.013, χ2 test). Participants randomized to group 2 notified the study coordinator when breakthrough bleeding/spotting episodes exceeded 5 days and were given specific instructions as to dates of removal and insertion. In only 2 (4%) of 50 months with interventions for breakthrough bleeding/spotting did participants not follow the instructions for replacement. In both situations, though, the total days of ring use did not exceed 31 days. No participants became pregnant while using the contraceptive ring.
During the baseline 21/7 phase, vaginal bleeding measured either as a proportion of days with bleeding/spotting or as the average daily bleeding score did not differ between groups (Table 1). Highest flows occurred during the 7-day hormone-free interval of the baseline 21/7 phase, with a subsequent reduction in flow noted with continuous transvaginal ring use (Fig. 1). Although the daily average flow scores were reduced throughout the 6-month continuous use interval, there was a recurring pattern of increased incidence of breakthrough bleeding/spotting surrounding the day of ring replacement each month. As can be seen in Figure 1, there is a small increase in average daily flow score approximately every 30 days (the average monthly calendar length). A flow score of 1 represents spotting only; as can be seen in the figure, flow was spotting only in most participants.
Because days in each calendar month vary from 28 to 31, results were examined on a daily basis from 1 to 31 days of ring use (Fig. 2). This figure shows the percentage of days that bleeding/spotting episodes began, divided by the total number of days studied for each day of ring use. For example, there were 420 first days of ring use (74 participants times 6 months [444 days] minus 24 days to account for participants discontinuing at various points). Only 3.5% of participants experienced the beginning of an episode of breakthrough bleeding/spotting on one of these 420 first days of ring use. Breakthrough bleeding/spotting initiated on the first day of new ring use was higher than during days 4 through 16. After the 16th day of continuous ring use, the number of times breakthrough bleeding/spotting was initiated increased, with the largest percentage of breakthrough bleeding/spotting initiating on day 29 (4.5%).
Six (8%) of 74 women had no bleeding or spotting episodes during the entire 6-month continuous phase of ring use, and 14 (19%) women had only spotting episodes. The median time to first breakthrough spotting episode was 48 days and to first breakthrough bleeding episode was 71 days of continuous contraceptive ring use with monthly changes. The pattern of daily flow scores on the scale from 0 to 4 for 74 participants is shown in Figure 3, with higher average flow scores on the first day of each new transvaginal ring and beginning again between days 18 and 22 of continuous ring use, peaking on day 29. Even in the 21/7 cycle, some increase in flow score is detectable during days 20 and 21 of ring use. For these reasons, the usual method of initial analysis of bleeding comparisons using 90-day intervals24 was not pursued. Instead, a secondary analysis method focusing on months with average flow scores and percentages of flow-free days was used, in keeping with recommendations for unusual situations.24
Table 2 shows comparisons within groups to detect differences in the percentage of flow-free days and average daily flow scores between the 21/7 cycle and monthly averages for participants who completed 6 months of continuous transvaginal contraceptive ring use. Twenty-nine participants in group 1 (without intervention for breakthrough bleeding/spotting) had complete data for the study, as did 36 participants in group 2 (4-day hormone-free interval for episodes of breakthrough bleeding/spotting of greater than 5 days). In group 1, percentage of flow-free days did not differ from the 21/7 baseline compared with the continuous phase (P=.086), but the average daily flow score was reduced with continuous use (P=.012). In group 2, there was an improvement in the bleeding profile, with an increase in flow-free days in the continuous phase compared with the baseline 21/7 phase (P<.001), and also a reduction in average daily flow scores (P=.001). Also seen in Table 2, group 2 had a greater percentage of flow-free days than did group 1 (P=.016) but no difference in average daily flow scores (P=.38).
Sixty-five participants completed 6 months, or an average of 181 days, of continuous ring use (P<.02). These 65 participants had a median of 7 days, or 1.2 days per month, of bleeding (average of 11 days or 1.8 days per month) and a median of 11 days, or 1.8 days per month, of spotting (average of 15 days or 2.5 days per month) during the entire 6-month continuous use interval compared with a median of 3 days of bleeding (average of 3.3 days) and 1 day of spotting (average of 1.8 days) during the 28 days of the baseline 21/7 cycle.
The effect on flow pattern of initiating a 4-day hormone-free interval also was examined by comparing flow scores between participants randomly assigned to group 1 (no intervention) and group 2 (4-day hormone-free interval for breakthrough bleeding/spotting episodes greater than 5 days’ duration). For this comparison of flow scores, 108 episodes of breakthrough bleeding/spotting greater than 5 consecutive days treated either with temporary removal of the contraceptive ring (4-day hormone-free interval) or no removal were examined using an intent-to-treat model. Fifty-eight of these episodes occurred in participants assigned to group 1. However, on one occasion, a participant did remove the ring for a hormone-free interval despite being told not to do so. Fifty episodes occurred in participants assigned to group 2, who were to remove the ring for a 4-day hormone-free interval. In seven cases, the removal was not performed because participants felt their flow was or would be decreasing. Figure 4 provides the results of the intent-to-treat analysis. The groups responded differently (P<.001) over the sampling interval from 5 days before the intervention to 14 day after the intervention using ANOVA. Removal of the ring resulted in higher flow scores for days 1, 2, and 3 of the hormone-free interval and lower flow scores beginning 8 days after the start of the hormone-free interval. As demonstrated in Figure 2, the majority of episodes of bleeding/spotting regardless of duration began in the second half of the month of continuous ring use, so, typically, the old ring would be removed and the new ring inserted within a few days of participants meeting criteria for the intervention.
After the sixth month of continuous contraceptive ring use, participants were allowed to continue for 6 additional months of continuous ring use. If they elected to continue, there were asked to select the method of intervention most desired for treatment of breakthrough bleeding/spotting lasting more than 5 days. Of 74 participants initiating 6 months of continuous contraceptive ring use, 65 (88%) completed the interval. Eight of the 65 elected to discontinue ring use after 6 months of continuous use, and the remaining 57 continued ring use. Of these 57, 37 (65%, 95% confidence interval 51–77%) chose to use no removal for treatment of breakthrough bleeding/spotting during another 6-month interval. The selection choices did not differ between groups (P=.17, χ2 test).
As with flow scores, continuous use of the contraceptive ring results in the absence of the usual cyclic increase in pelvic pain, headache, and mood lability associated with the 7-day hormone-free interval. Although a slight increase in flow score was seen around the time of ring replacement during continuous use (Fig. 1), an increase in symptoms was not seen. For pelvic pain, average scores did not differ (P=.77) between groups but did differ (P=.001) between initial 21/7 cycle (0.32±0.05) and the months of continuous use (0.18±0.03). For headache, average scores did not differ (P=.19) between groups and tended to vary (P=.066) between initial 21/7 cycle (0.45±0.05) and the months of continuous use (0.32±0.06). For mood, average scores did not differ (P=.19) between groups and did not differ (P=.73) between initial 21/7 cycle (1.76±0.28) and the months of continuous use (1.24±0.28). Although statistically significant reductions in headache and mood scores were not seen with continuous ring use, it is important to note that patients with cyclic symptoms were not specifically recruited and the study was not powered for that purpose.
Combination (estrogen plus progestin) contraception has undergone major changes in the past 5 years, with all regimens newly approved by the U.S. Food and Drug Administration deviating from the standard 21/7 design. Modifications have included an extended OC regimen (84/7) (Seasonale [Duramed Pharmaceuticals, Montvale, NJ]),13 shortening of the typical 7-day hormone-free interval with the introduction of 24/4 regimens (Yaz [Bayer Healthcare Pharmaceuticals, Wayne, NJ] and Loestrin 24 [Warner Chilcott, Rockaway, NJ]), the addition of low-dose estrogen to the typical 7-day hormone-free interval of an extended regimen (84/7 ethinyl estradiol, Seasonique [Duramed Pharmaceuticals]),12 and a continuous combined regimen (Lybrel [Wyeth, Madison, NJ]).14 Although extended and continuous OC regimens have become available only recently, off-label extended use of the 21/7 regimen has been commonplace in clinical practice for treatment of endometriosis, prevention of hormone-withdrawal symptoms, and convenience.17 The most common side effect with extended use is breakthrough bleeding/spotting. Retrospective and prospective data from our institution have documented successful resolution of this nuisance problem by instituting an abbreviated 3- to 4-day hormone-free interval with immediate resumption of active pills.10,11 Continuous and extended use of the vaginal ring also has been reported21,22 and is commonly employed by clinicians in various regimen designs.
Our study design incorporated a monthly (28 to 31 calendar day) replacement of the vaginal ring in current users of combination hormonal contraception with normal pelvic sonography at low risk for anatomic causes of breakthrough bleeding/spotting. Use of this continuous method resulted in a significant reduction in flow compared with the baseline 21/7 cycle. Continuation rate was good, with the majority of participants electing use of the continuous ring after completion of the 6-month trial. As with continuous OC regimens, breakthrough bleeding/spotting was noted. Interestingly, occurrence of breakthrough bleeding/spotting peaked around the time of ring replacement (Fig. 1), appearing in a small percentage of participants as early as day 17 of ring use (Fig. 2). As seen in Figure 3, the longer the calendar month, the greater the flow before ring replacement. As previously demonstrated by Timmer and Mulders, the vaginal contraceptive ring delivers a peak of circulating hormone at about 1 week after inserting a new ring, followed by a gradual linear decrease with time out to 35 days.23 It is possible that in some patients, the gradually decreasing hormones allow for breakthrough bleeding/spotting before the end of the monthly duration. In addition, once a ring is inserted, peak serum hormone levels are not reached until about 1 week, which may further explain the slight increase in breakthrough bleeding/spotting in some patients around the time of ring exchange. Although breakthrough bleeding/spotting greatly decreases over time with continuous oral contractive pill use,11–16 this was not seen with monthly replacement of the ring because of the slight increase of breakthrough bleeding/spotting seen around the time of monthly ring replacement (Figs. 1, 2, and 3).
Overall, the bleeding profile was good. Although most participants experienced some breakthrough bleeding/spotting as expected, six of 65 participants who completed the 6 months of continuous use had no breakthrough bleeding or breakthrough spotting and 14 had only breakthrough spotting and no breakthrough bleeding. Group 2, which was allowed to remove the ring for 4 days if breakthrough bleeding/spotting developed, experienced a statistically greater percentage of days without breakthrough bleeding/spotting (95%) compared with group 1 (89%), which had no hormone-free days (Table 2). Interestingly, after 6 months of continuous use, the majority of those who wished to continue ring use said they would prefer not to remove the ring and reinsert it for breakthrough bleeding/spotting. Obviously, discontinuing OCs for 4 days for breakthrough bleeding/spotting is not as cumbersome as removing, storing, and reinserting a vaginal ring. Nonetheless, about a third of the group stated they would prefer this as a method of management of bleeding.
Similar to studies of continuous OCs,18–20 continuous ring use resulted in a reduction not only in flow but also in hormone-withdrawal symptoms. With the power provided in this study, the pelvic-pain score was reduced significantly, whereas headache and mood scores were not significantly reduced compared with the baseline 21/7 cycle. Participants with hormone-withdrawal symptoms were not specifically selected for this trial, so the changes related to the 7-day hormone-free interval for headache and mood were not increased during the baseline 21/7 phase sufficiently enough to allow detection. Many patients discontinue the pill, patch, and ring because of side effects that may be induced artificially during the 7-day hormone-free interval of these standard 21/7 regimens. All patients on a 21/7 regimen should be queried regarding hormone-withdrawal symptoms, and alternative regimens should be discussed if cyclic symptoms are noted.
Continuous vaginal ring use resulted in an acceptable bleeding profile in most patients, reduction in flow, reduction in pelvic pain, and a high continuation rate. Patients desiring to initiate a continuous vaginal ring regimen should be counseled on the possibility of breakthrough bleeding/spotting. Although many will have minimal to no breakthrough bleeding/spotting with monthly calendar replacement, patients should be instructed to shorten the duration of ring use to 3 to 4 weeks if bothersome bleeding occurs. Removal for 4 days with reinsertion of the same ring can improve the bleeding profile, with fewer days of breakthrough bleeding/spotting.
1. Bäckström T, Hansson-Malmström Y, Lindhe BÅ, Cavalli-Björkman B, Nordenström S. Oral contraceptives in premenstrual syndrome: a randomized comparison of triphasic and monophasic preparations. Contraception 1992;46:253–68.
2. Joffe H, Cohen LS, Harlow BL. Impact of oral contraceptive pill use on premenstrual mood: predictors of improvement and deterioration. Am J Obstet Gynecol 2003;189:1523–30.
3. Sulak PJ, Scow RD, Preece C, Riggs MW, Kuehl TJ. Hormone withdrawal symptoms in oral contraceptive users. Obstet Gynecol 2000;95:261–6.
4. Spona J, Elstein M, Feichtinger W, Sullivan H, Lüdicke F, Müller, U, et al. Shorter pill-free interval in combined oral contraceptives decreases follicular development. Contraception 1996;54:71–7.
5. Sullivan H, Furniss H, Spona J, Elstein M. Effect of 21-day and 24-day oral contraceptive regimens containing gestodene (60 μg) and ethinyl estradiol (15 μg) on ovarian activity. Fertil Steril 1999;72:115–20.
6. Killick SR, Fitzgerald C, Davis A. Ovarian activity in women taking an oral contraceptive containing 20 microg ethinyl estradiol and 150 microg desogestrel: effects of low estrogen doses during the hormone-free interval. Am J Obstet Gynecol 1998;179:S18–24.
7. Schlaff WD, Lynch AM, Hughes HD, Cedars MI, Smith DL. Manipulation of the pill-free interval in oral contraceptive pill users: the effects on follicular suppression. Am J Obstet Gynecol 2004;190:943–51.
8. Willis SA, Kuehl TJ, Spiekerman AM, Sulak PJ. Greater inhibition of the pituitary–ovarian axis in oral contraceptive regimens with a shortened hormone-free interval. Contraception 2006;74:100–3.
9. Holt VL, Cushing-Haugen KL, Daling JR. Oral contraceptives, tubal sterilization, and functional ovarian cyst risk. Obstet Gynecol 2003;102:252–8.
10. Sulak PJ, Carl J, Gopalakrishnan I, Coffee A, Kuehl TJ. Outcomes of extended oral contraceptive regimens with a shortened hormone-free interval to manage breakthrough bleeding. Contraception 2004;70:281–7.
11. Sulak PJ, Kuehl TJ, Coffee A, Willis S. Prospective analysis of occurrence and management of breakthrough bleeding during an extended oral contraceptive regimen. Am J Obstet Gynecol 2006;195:935–41.
12. Anderson FD, Gibbons W, Portman D. Safety and efficacy of an extended-regimen oral contraceptive utilizing continuous low-dose ethinyl estradiol. Contraception 2006;73:229–34.
13. Anderson FD, Hait H. A multicenter, randomized study of an extended cycle oral contraceptive [published erratum appears in Contraception 2004;69:175]. Contraception 2003;68:89–96.
14. Archer DF, Jensen JT, Johnson JV, Borisute H, Grubb GS, Constantine GD. Evaluation of a continuous regimen of levonorgestrel/ethinyl estradiol: phase 3 study results. Contraception 2006;74:439–45.
15. Kwiecien M, Edelman A, Nichols MD, Jensen JT. Bleeding patterns and patient acceptability of standard or continuous dosing regimens of a low-dose oral contraceptive: a randomized trial. Contraception 2003;67:9–13.
16. Miller L, Hughes JP. Continuous combination oral contraceptive pills to eliminate withdrawal bleeding: a randomized trial. Obstet Gynecol 2003;101:653–61.
17. Sulak P, Buckley T, Kuehl TJ. Attitudes and prescribing preferences of health care professionals in the United States regarding use of extended-cycle oral contraceptives. Contraception 2006;73:41–5.
18. Coffee AL, Kuehl TJ, Willis S, Sulak PJ. Oral contraceptives and premenstrual symptoms: comparison of a 21/7 and extended regimen. Am J Obstet Gynecol 2006;195:1311–9.
19. Sulak P, Willis S, Kuehl T, Coffee A, Clark J. Headaches and oral contraceptives: Impact of eliminating the standard 7-day placebo interval. Headache 2007;47:27–37.
20. Coffee AL, Sulak PJ, Kuehl TJ. Long-term assessment of symptomatology and satisfaction of an extended oral contraceptive regimen. Contraception 2007;75:444–9.
21. Barreiros FA, Guazzelli CA, deAraújo FF, Barbosa R. Bleeding patterns of women using extended regimens of the contraceptive vaginal ring. Contraception 2007;75:204–8.
22. Miller L, Verhoeven CH, Hout J. Extended regimens of the contraceptive vaginal ring: a randomized trial. Obstet Gynecol 2005;106:473–82.
23. Timmer CJ, Mulders TM. Pharmacokinetics of etonogestrel and ethinylestradiol released from a combined contraceptive vaginal ring. Clin Pharmacokinet 2000;39:233–42.
24. Gerlinger C, Endrikat J, Kallischnigg G, Wessel J. Evaluation of menstrual bleeding patterns: a new proposal for a universal guideline based on the analysis of more than 4500 bleeding diaries. Eur J Contracept Reprod Health Care 2007;12:203–11.
25. Coffee AL, Kuehl T, Sulak P. Comparison of scales for evaluating premenstrual symptoms in women using oral contraceptives. Pharmacotherapy 2008;28:576–83.