Continuation Rate of Combined Hormonal Contraception: A Prospective Multicenter Study

Abstract

Background:

Data from clinical trials regarding continuation of use and contraceptive efficacy do not always apply to the general public. Therefore, a study among typical users was conducted to assess the continuation rate at the end of 12 cycles of combined hormonal contraceptive methods, reasons for discontinuation, and the Pearl index.

Methods:

Prospective, observational, and multicenter study of 3443 women aged 18 to 49 years starting one of the three combined hormonal contraception methods available in Spain (the vaginal ring, the contraceptive pill, and the transdermal skin patch).

Results:

The study population (intention-to-treat analysis) included 3443 women, of whom 45.4% were included in the vaginal ring group, 42.6% the pill group, and 12.1% the skin patch group. The continuation rate at 12 cycles was 45.9% for the pill, 42.3% for the vaginal ring, and 26.0% for the skin patch. The Pearl index was 0.61 (95% confidence interval [CI] 0–1.2) for the pill, 0.61 (95% CI 0–1.1) for the vaginal ring, and 2.34 (95% CI 0.3–9) for the skin patch (p<0.001).

Conclusion:

At 12 cycles, the vaginal ring and the pill showed similar continuation rates and effectiveness, which were significantly higher than the skin patch.

Introduction

At least three combined hormonal contraceptive methods, the vaginal ring, the pill, and the transdermal skin patch, are available in many countries of the world. Once-a-month estrogen–progesterone injectables are also available in some countries. The theoretical efficacy of oral combined contraception is very high, with failure rates of 0.12 to 0.34 pregnancies per 100 women-years for perfect use in controlled clinical trials.^1,2 In clinical practice, failure rates of about 3%–8% per year in the general population during the first year of use are found,³ with even higher rates up to 16%–20% per year in populations at risk.⁴ This discrepancy between use-effectiveness and theoretical effectiveness is likely a result of the need for daily compliance when using oral contraceptives.

The development of new nondaily combined hormonal contraceptive methods can potentially improved compliance, and initial clinical trials comparing the contraceptive patch to oral contraceptives showed significantly better compliance among users of the contraceptive patch.⁵ However, comparative studies of the vaginal ring and the pill did not show differences in compliance between the two methods.⁶ A recent Cochrane review demonstrated similar effectiveness for the three methods.⁷ Conclusions of these studies are based on data obtained from clinical trials in which participants followed a protocol and were seen regularly, so it is likely that results in terms of compliance cannot be applied to the general population. A cross-sectional multicenter study among 26,250 typical users of a combined hormonal contraceptive method showed statistically significant differences in noncompliant behavior, with the largest percentage among pill users (71%), followed by patch users (32%) and vaginal ring users (21.6%).⁸ A limitation of this study is that information was collected retrospectively. In order to avoid this source of bias, a prospective study was designed, with the main objective being to determine the percentage of women still using a combined hormonal contraceptive method after 12 cycles of use. Secondary objectives were to assess the main reasons for early discontinuation and the Pearl index after 12 cycles.

Materials and Methods

Design and study population

A prospective observational multicenter study was designed to assess the continuation rate at the end of 12 cycles for each of the three combined hormonal contraception methods available in Spain. These included the vaginal ring, the contraceptive pill, and the transdermal skin patch. Secondary objectives of the study were to identify the reasons for discontinuation of each method and to determine the Pearl index for typical users after 1 year of use. The study was carried out at the outpatient clinics of the departments of gynecology and family planning centers of public and private institutions throughout Spain in routine clinical settings. None of the three combined hormonal contraceptive methods is reimbursed by the Spanish National Health System. All three medicines have similar retail prices and are dispensed in official pharmacies of the General Direction of Pharmacies and Sanitary Products (Ministry of Health, Social Policy and Equality) with a medical prescription. Combined hormonal contraceptive drugs are usually prescribed for 12 months of use. The study protocol was approved by the Ethics Committee on Clinical Research of the Autonomous Community of the Basque Country.

Between June 1, 2007, and May 31, 2008, all consecutive women aged 18 to 49 years old who consulted their doctors for contraception and started combined hormonal contraception were eligible, provided that they gave written informed consent and were able to understand and complete the study procedures. Women with contraindications for the use of combined hormonal contraception according to the package insert specifications of the products, eligibility criteria of the World Health Organization (WHO), and contraindications established by the Spanish Society of Contraception were excluded as were those who selected a contraceptive method other than combined hormonal contraception. Women who had previously used a combined hormonal contraceptive method were also excluded. Participating physicians were selected among members of the Spanish Society of Contraception who had been involved in a previous study (TEAM study) designed to gather information on the reasons for selecting the contraceptive pill, the skin patch, and the vaginal ring.⁹ During the performance of the TEAM study, the 100 most active physicians in the recruitment of patients and collection of data were identified. All these physicians were invited to take part in this study and 92 accepted.

Study procedures and data collection

At the initial visit, all women who wished to use hormonal contraception received contraceptive counseling including a full discussion of the effectiveness of the oral pill, the vaginal ring, and the contraceptive skin patch as well as the related modes of action, modes of use, risks and benefits, and suitability for individual needs. All investigators used the same leaflet to back up verbal information about the combined hormonal contraceptive methods, a copy of which was given to each women at the time of consultation to facilitate contraception choice. The leaflet contained salient information on the mechanism of action of each method, efficacy, usage, route, frequency of use, and other important characteristics. The participants also completed a self-administered questionnaire regarding contraception choice. Details of this questionnaire have been previously reported.⁹ Briefly, a list of reasons for the preference of each category (pill, skin patch, vaginal ring) (e.g., frequency of use, convenience, lower probability of inadvertent omission, easy handling, proven efficacy) and a list of reasons for not choosing the remaining methods (e.g., suspicion of inconvenience, distrust of efficacy, partner's refusal, fear of detachment [skin patch], fear to wear a foreign body [vaginal ring], need of daily intake [contraceptive pill], acquaintances' contrary opinion) were included in the questionnaire, with multiple choice answers in all categories.

Before prescription of the contraceptive method, women signed the informed consent to take part in the study and a written consent to allow an external clinical research organization to make a telephone call at 12 cycles from the prescription date. All participants were appointed for a follow-up visit after 12 cycles of use. Women who failed to attend the control visit were contacted by phone. The follow-up period ended on May 31, 2009.

For each woman included in the study, the physician recorded basic demographic data, the contraceptive method currently used, and the contraceptive method selected after counseling (contraceptive pill, skin patch, vaginal ring). At the final visit, data related to continuation of use of the contraceptive method selected and reasons for discontinuation and data referring to failures of the method were collected using a simple questionnaire, specifically designed for that purpose.

Sample size and statistical analysis

Calculation of the sample size was based on the main objective and primary endpoint of the study; i.e., to determine the percentage of women who continue to use one of the three combined hormonal contraceptive methods at 12 cycles after initiation. Based on a previous study¹⁰ in which 68% women who started combined hormonal contraception for the first time continued using the method at six cycles after initiation, and according to binomial distribution, a sample size of 3672 women would provide an accuracy of±1.5% with a confidence level of 95%.

Intention-to-treat analysis included all women who met the inclusion criteria and for which data for the primary endpoint of the study were recorded. The safety data set used for the description of baseline characteristics and adverse events included all women who used one of the combined hormonal contraception methods. Categorical variables are expressed as absolute frequencies and percentages, and continuous variables as mean and standard deviation (SD). The chi-square (χ²) test was used for the comparison of categorical variables. Statistical significance was set at p<0.05.

Results

During the recruitment period, 3535 women were eligible but 92 (2.6%) were excluded because of inconsistencies during the process of data validation. Therefore, the intention-to-treat population consisted of 3443 women, 1562 of which were included in the group of the vaginal ring, 1465 in the group of the contraceptive pill, and 416 in the group of the skin patch. However, 509 (14.8%) women did not start the contraceptive method prescribed, including 230 (14.7%) of those who selected the vaginal ring, 199 (13.6%) the pill, and 80 (19.2%) the skin patch.

The remaining 2934 patients started their prescribed method. The mean (SD) age of these women was 26.2 (6.2) years. Of these women, 45.4% (n=1332) began using the ring, 43.1% (n=1266) the pill, and 11.4% (n=336) the transdermal skin patch. Most women (94.2%) had a stable partner, 85.6% of the participants had a university degree or secondary education, and 65.9% were employed out of home. Previous contraceptive methods included condoms for 83.7% of women, withdrawal for 5.5%, nonhormonal intrauterine device for 1.5%, other methods for 2.9%, and no prior contraception for 6.4%.

The flow-chart of the study population is shown in Fig. 1. During the follow-up period of 12 cycles, 496 (16.9%) women were lost to follow-up and 997 (33.9%) discontinued treatment. A significantly lower percentage of women preferring the vaginal ring (35.1%) or the pill (27.6%) discontinued the contraception method as compared with those preferring the skin patch (53.3%) (p<0.001). Reasons given for abandonment of treatment are detailed in Table 1. A similar percentage of women in the group of each hormonal contraceptive method argued poor cycle control, tolerability, the desire to get pregnant, or a change in sexual habits. Failure of the contraceptive method was reported by 2.2% of women using the skin patch as compared with 1.4% and 1.1% of those using the pill and the vaginal ring, respectively (p<0.001). In addition, “other reason” was also more commonly recorded among skin patch users (36.3%) and vaginal ring users (30.1%) than among pill users (16.3%) (p<0.001).

FIG. 1.

Flow-chart of the study population.

Table 1.

Reasons for Discontinuation of the Combined Hormonal Contraceptive Method in 2934 Women Who Started Their Prescribed Method

Reason	Vaginal ring, n (%)	Pill, n (%)	Skin patch, n (%)
Poor cycle control	56 (11.9)	44 (12.5)	14 (7.8)
Side effects	151 (32.1)	161 (46.1)	67 (37.5)
Pregnancy desire	40 (8.5)	29 (8.3)	15 (8.4)
Change in sexual habits	73 (15.9)	54 (15.4)	14 (7.8)
Failure of the method^a	5 (1.1) a	5 (1.4) a	4 (2.2) b
Other^a	143 (30.6) c	57 (16.3) d	65 (36.3) c
Total	468	350	179

a versus b and c versus d, p<0.001.

The continuation rate after 12 cycles was 42.3% (n=661/1562) for the vaginal ring, 49.5% (n=672/1465) for the pill, and 26.0% (n=108/416) for the skin patch. In patients who started their prescribed method (Fig. 2), the continuation rates were 49.6% (n=661/1332) for the vaginal ring, 53.1% (672/1266) for the pill, and 32.1% (108/336) for the skin patch. The proportions of women who discontinued their prescribed method were 35.1% (n=468/1332) for the vaginal ring, 27.6% (n=350/1266) for the pill, and 53.3% (n=179/336) for the skin patch. Differences between the skin patch and the other two methods were statistically significant (p<0.001).

FIG. 2.

Percentages of women who discontinued treatment and the continuation rate at the end of 12 cycles for each of the three combined hormonal contraceptive methods in patients who started their prescribed method.

The method that had the highest proportion of women who decided to change was the skin patch (12%), followed by the vaginal ring (6.1%) and the pill (3.5%). There were statistically significant differences between change of treatment and the method selected at the baseline visit (χ², p<0.05). Users of the vaginal ring and the skin patch mainly chose the contraceptive pill (50.5% and 70%, respectively) and among users of the pill, 48.1% chose the vaginal ring.

Fourteen women became pregnant: five among vaginal ring users, five among pill users, and four among skin patch users. Vaginal ring users completed 10,725 treatment cycles, pill users 10,706, and skin patch users 2228. The estimated Pearl index for each combined hormonal contraceptive method was 0.61 (95% CI 0–1.1) for the vaginal ring, 0.61 (95% CI 0–1.2) for the contraceptive pill, and 2.34 (95% CI 0–3.9) for the skin patch. There were statistically significant differences in the Pearl index for the skin patch (2.34) as compared with the pill (0.61) and the vaginal ring (0.61) (p<0.001).

The number of women for each method who got a prescription, started the method, and continued for 1 year, and the total number of cycles and pregnancies are summarized in Table 2.

Table 2.

Summary of Study Findings

	Vaginal ring, n (%)	Contraceptive pill, n (%)	Skin patch, n (%)
Intention-to-treat population, n=3443	1562	1465	416
Continuation rate at 1 year	661 (42.3)	672 (45.9)	108 (26.0)^*
Started the method, n=2934	1332	1266	336
Continuation rate at 1 year	661 (49.6)	672 (53.1)	108 (32.1)
Total no. of cycles	10,725	10,706	2228
No. of pregnancies	5	5	4

p<0.001.

Discussion

We report the results of a study carried out in typical users of combined hormonal contraception. To our knowledge, no previous study aimed at determining the continuation rate of combined hormonal contraception methods in more than 3000 women followed prospectively for 12 cycles of contraceptive use has been published in the literature. A total of 3443 women were recruited, but 509 (14.8%) did not start using the contraceptive method prescribed at the initial visit. This proportion is much higher than observed in some clinical trials in which the number of abandonments after randomization and before starting treatment was quantified, with a rate of 3% for the vaginal ring in a clinical trial of more than 100 women¹¹ and a rate of 5% for the skin patch in a other comparative study of about 1500 participants.¹² This finding should draw attention to the quality of contraceptive counseling and the role played by professionals in the final decision of the users. It is likely that in some cases, criteria may be imposed by professionals so that the final choice made does not encompass the woman's real needs. This may be the reason why some of them refused to initiate a new contraceptive method and continued to use, in most cases, the method that they had been using. In this respect, it is important to remember that complete and unbiased information of the different contraception choices should be provided to assist the user to make free and informed decisions.¹³ Besides contraceptive counseling and motivation, this study was conducted in conditions of daily practice among typical users to which free contraception was not provided.

The overall rate of loss to follow-up among women who started treatment was 16.9%, which is somewhat lower than proportions ranging between 21% and 37% reported in clinical trials with large study samples.^11

–15 It is possible that the methodology used in our study in which women gave consent to receive a telephone call after 12 cycles of use may had an influence on the lower rate of loss to follow-up.

One out of three women who started the use of a combined hormonal contraceptive method abandoned its use during the first year, with a particularly higher incidence for the transdermal skin patch (53.3%). In the case of the contraceptive pill, the discontinuation rate of 27.6% found in our population is similar to that reported in the United States based on a population drawn from the 2002 National Survey of Family Growth in which oral contraceptives were associated with a risk of discontinuation of 29% during the first 6 months of use.¹⁶ A total of 35.1% of women using the vaginal ring discontinued its use within the first year of follow-up. In a randomized controlled trial of 136 women assigned to the contraceptive vaginal ring and 137 to the low-dose oral contraceptive pill, only 26% of contraceptive vaginal ring users and 29% of contraceptive pill users continued their assigned study method at 6 months.¹⁷ Reasons for discontinuation, however, were not reported. It is well known that the appearance of side effects during oral contraception is associated with a twofold relative risk for discontinuation.⁹ In the present study, side effects were the main reason for combined hormonal contraception discontinuation independent of the method prescribed, with a higher proportion of users of the contraceptive pill, in agreement with results of the study of Sabatini and Cagiano,¹⁸ although differences were not statistically significant. It is possible that a more extensive discussion of side effects at the time of consultation may contribute to reduce the number of abandonments.

The continuation rate at 12 cycles was 42.3% for the vaginal ring, 45.9% for the oral contraceptive pill, and 26.0% for the skin patch. Differences between the vaginal ring and the pill did not reach statistical significance, but the continuation among skin patch users was significantly lower than for the other two methods. Our findings are in agreement with data reported by others. In a recent 12-month longitudinal cohort study of 1387 adolescent girls and women, aged 15 to 24 years, continuation rates at 12 months among ring initiators and pill initiators was comparable, 29.4 and 32.7 per 100 person-years, respectively; however, the continuation rate was lowest for skin patch and depot medroxyprogesterone initiators, 10.9 and 12.1 per 100 person-years, respectively (p≤0.003).¹⁹ The low rate of continuation associated with the use of the skin patch is similar to that reported in other studies^5,13,15 and is particularly attributable to side effects,^20,21 which in the case of the skin patch were related to mastodynia and detachment.

A secondary objective of the study was to determine the Pearl index for typical users of combined hormonal contraceptive methods. Pearl indices were 0.61 for the vaginal ring, 0.61 for the contraceptive pill, and 2.34 for the skin patch. These results suggest that among typical users, the vaginal ring and the pill have a similar effectiveness, whereas the effectiveness of the skin patch was lower. The present results are consistent with data obtained in clinical trials for the use of oral contraceptives and the vaginal ring under ideal conditions^6,12,14 but disagree with effectiveness data reported by Trussell²² that indicated 8% of women using combined hormonal contraception will experience an unwanted pregnancy during the first year of use. This important difference in the effectiveness rate, 8% in the study of Trussell²² and 0.5% in the present study, may be explained by differences in the sociodemographic characteristics of women included in this study and those of women in the National Survey Family Growth cohort. In addition, it is possible that effectiveness indices may vary if information on the 469 participants who were lost at follow-up would have been available. Despite this fact, effectiveness of the vaginal ring and the contraceptive pill in our study are close to the efficacy under ideal conditions of use reported in clinical trials.

In the case of the skin patch, different results were obtained. According to data from clinical trials, the Pearl index for this method varies between 0.71² to 1.24³ but was higher (2.34) in our study. It has been suggested that efficacy of the skin patch is lower in obese women weighing more than 90 kg,²³ but none of the four women who became pregnant while using this method of contraception weighed more than 90 kg. The mean body mass index in these women was 22.8 (3.7) kg/m².

Our study has some limitations. This was not a randomized clinical trial but a prospective study in which women selected the contraceptive method that best suited their needs. This design, intended to reflect the behavior of typical users, allowed the introduction of a selection bias, so that it may have been possible that women who chose the transdermal skin patch and presented a worse continuation rate and a worse Pearl index had a different baseline sociodemographic characteristics. Moreover, while we were able to collect information on more than 10,000 cycles for the groups of the vaginal ring and the contraceptive pill, information on only 2228 cycles was available in the skin patch group, which may have affected the results obtained.

Finally, the difference between the calculated sample size of 3672 women and 3535 eligible women may be explained by the restrictive entry criterion; that is, women who were starting combined hormonal contraception for the first time and had no previous experience with combined hormonal treatment as a birth control method. However, sociodemographic data of the study sample are similar to the general Spanish population and consistent with findings reported in surveys of the use of contraceptive methods in Spain.²⁴ On the other hand, the skin patch group was only one third the size of the other two groups because only that number of women chose that method. This proportion is similar to data in the TEAM study⁹ and reflects the preferences of Spanish women regarding the use of this type of contraception.

In summary, the present findings indicate that among typical users of combined hormonal contraception methods, the oral contraceptive pill and the vaginal ring had similar results regarding continuation rates at the end of 12 cycles and effectiveness, whereas the continuation rate for 12 cycles for the transdermal skin patch was significantly lower. There is a need to provide complete and adequate information during contraceptive counseling to reduce the number of women who discontinue contraception due to side effects, to improve the continuation rate of combined hormonal contraception, and to decrease the number of unintended pregnancies.

Footnotes

Acknowledgments

We thank Marta Pulido, M.D., for editing the manuscript and for editorial assistance. This study was conducted with an educational grant from Schering Plough España, S.A., Madrid, Spain.

Author Disclosure Statement

I. Lete, E. Pérez-Campos, E. de la Viuda, N. Mendoza, J. González, C. Castelo-Branco, and P. Lobo have fees for talks in sponsored symposia of Bayer, MSD, and Grunenthal.

References

International Working Group on Enhancing Patient Compliance and Oral Contraceptive Efficacy. A consensus statement: enhancing patient compliance and oral contraceptive efficacy. Br J Fam Plann, 1993; 18:126–129.

Potter

, Oakley

, de Leon-Wong

, Cañamar

. Measuring compliance among oral contraceptive users. Fam Plann Perspect, 1996; 28:154–158.

Rosenberg

, Waugh

, Burnhill

. Compliance, counseling and satisfaction with oral contraceptives. A prospective evaluation. Fam Plann Perspect, 1998; 30:89–104.

Wahab

, Killick

. Oral contraceptive knowledge and compliance in young women. Br J Fam Plann, 1996; 22:170–173.

Dittrich

, Parker

, Rosen

, Shangold

, Creasy

, Fisher

. Transdermal contraception: evaluation of three transdermal norelgestromin/ ethinyl estradiol doses in a randomized, multicenter, dose-response study. Am J Obstet Gynecol, 2002; 186:15–20.

Milsom

, Lete

, Bjertnaes

et al. Effects on cycle control and bodyweight of the combined contraceptive ring, NuvaRing, versus an oral contraceptive containing 30 microg ethinyl estradiol and 3 mg drospirenone. Hum Reprod, 2006; 21:2304–2311.

Lopez

, Grimes

, Gallo

, Schulz

. Skin patch and vaginal ring versus combined oral contraceptives for contraception. Cochrane Database of Systematic Reviews, 2010Issue 3 Art. No: CD003552 DOI:10.1002/14651858.

Lete

, Doval

, Pérez-Campos

et al. Self-described impact of noncompliance among users of a combined hormonal contraceptive method. Contraception, 2008; 77:276–282.

Lete

, Doval

, Pérez-Campos

et al. Factors affecting women's selection of a combined hormonal contraceptive method: the TEAM-06 Spanish cross-sectional study. Contraception, 2007; 76:77–83.

10.

Rosenberg

, Waugh

. Oral contraceptive discontinuation: a prospective evaluation of frequency and reasons. Am J Obstet Gynecol, 1998; 179:577–582.

11.

Ahrendt

H-J

, Nisand

, Bastianelli

et al. Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive containing 3 microg of ethinyl estradiol and 3 mg of drospirenone. Contraception, 2006; 74:451–457.

12.

Audet

, Moreau

, Koltun

et al. Evaluation of contraceptive efficacy and cycle control of a transdermal contraceptive patch vs an oral contraceptive: a randomized controlled trial. JAMA, 2001; 285:2347–2354.

13.

Lete

. What about compliance? Eur J Contracept Reprod Health Care, 2010; 15:217–219.

14.

Oddsson

, Leifels-Fischer

, de Melo

et al. Efficacy and safety of a contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive: a 1-year randomized trial. Contraception, 2005; 71:176–182.

15.

Urdl

, Apter

, Alperstein

et al. Contraceptive efficacy, compliance and beyond: factors related to satisfaction with once-weekly transdermal compared with oral contraception. Eur J Obstet Gynecol Reprod Biol, 2005; 121:202–210.

16.

Moreau

, Cleland

, Trussell

. Contraceptive discontinuation attributed to method dissatisfaction in the United States. Contraception, 2007; 76:267–272.

17.

Gilliam

, Neustadt

, Kozloski

, Mistretta

, Tilmon

, Godfrey

. Adherence and acceptability of the contraceptive ring compared with the pill among students. A randomized controlled trial. Obstet Gynecol, 2010; 115:503–510.

18.

Sabatini

, Cagiano

. Comparison profiles of cycle control, side effects and sexual satisfaction of three hormonal contraceptives. Contraception, 2006; 74:220–223.

19.

Raine

, Foster-Rosales

, Upadhyay

et al. One-year contraceptive continuation and pregnancy in adolescent girls and women initiating hormonal contraceptives. Obstet Gynecol, 2011; 117:363–371.

20.

Kluft

, Meijer

, LaGuardia

, Fisher

. Comparison of a transdermal contraceptive patch vs. oral contraceptives on hemostasis variables. Contraception, 2008; 77:77–83.

21.

Smallwood

, Meador

, Lenihan

, Shangold

, Fisher

, Creasy

. Efficacy and safety of a transdermal contraceptive system. Obstet Gynecol, 2001; 98:799–805.

22.

Trussell

. Contraceptive failure in the United States. Contraception, 2004; 70:89–96.

23.

Zieman

, Guillebaud

, Weisberg

, Shangold

, Fisher

, Creasy

. Contraceptive efficacy and cycle control with the Ortho Evra™/Evra™ transdermal system. Fertil Steril, 2002; 77:13–18.

24.

Dueñas

, Lete

, Bermejo

et al. Trends in the use of contraceptive methods and voluntary interruption of pregnancy in the Spanish population during 1997–2007. Contraception, 2011; 83:82–87.