Secondary Logo

Journal Logo

Contents: Original Research

Effects of Age, Parity, and Device Type on Complications and Discontinuation of Intrauterine Devices

Aoun, Joelle MD; Dines, Virginia A. BS; Stovall, Dale W. MD; Mete, Mihriye PhD; Nelson, Casey B. BA; Gomez-Lobo, Veronica MD

Author Information
doi: 10.1097/AOG.0000000000000144
  • Free

The most commonly chosen methods of contraception by adolescents in the United States are withdrawal, condoms, and oral contraceptive pills.1 These methods are associated with relatively high typical use failure rates and low rates of continuation, which, in turn, contribute to high rates of unplanned teen pregnancies.1 In fact, 82% of the approximately 750,000 adolescent pregnancies that occur every year in the United States are unintended.2 Of all reversible contraceptives, intrauterine devices (IUDs) have been shown to have one of the highest rates of satisfaction and continuation and an efficacy that is comparable to sterilization.3,4 The safety of IUD use in adolescents and nulliparous women is supported by recent recommendations from the U.S. Medical Eligibility Criteria for Contraceptive Use.5 However, IUDs remain underused in this patient population6 with only 4.5% of adolescents and women under the age of 20 years relying on long-acting reversible contraception.7 There is a lack of knowledge and some misconceptions among both patients and health care practitioners regarding IUD use in adolescents and nulliparous women.8–12 More specifically, because of the paucity of information regarding the use of IUDs in teenagers, health care practitioners remain hesitant to offer this method of contraception to this population primarily as a result of their concerns about its safety and efficacy.10–12

The purpose of this study was to conduct an analysis of IUD-related outcomes including expulsion, contraceptive failure, and early discontinuation and to compare these outcomes in regard to age, parity, and type of IUD used.


This study design was a multicenter retrospective chart review. Data were obtained from the electronic medical records of three independent medical centers. Of the three centers, one was Riverside Health System, a community-based institution with multiple sites in over 100 locations in southeastern Virginia. The second center was MedStar Health System, a large institution that includes 10 hospitals in the Baltimore–Washington metropolitan region. The third study center was Unity Health Care, one of the largest federally funded community health centers in the United States located in the Washington, DC, area. The study was approved by the institutional review boards of the three respective institutions. Data collection was performed by three individuals using a single standardized data sheet. This data sheet was beta tested initially on the first 40 patients with all three data collectors present and modified appropriately to ensure uniformity. Specific definitions regarding all clinical outcomes of interest were established before data collection and used consistently by all three collectors. Updates on data collection were done on a weekly basis, and regularly scheduled meetings were held to review the consistency the data collected.

Adolescents and women between the ages of 13 and 35 years who had undergone placement of either the levonorgestrel intrauterine system or the copper IUD specifically for the purpose of contraception between June 1, 2008, and June 30, 2011, were considered for inclusion in the study. Patients were identified by searching the electronic medical records of each center using the IUD-insertion procedure code. We chose not to include patients older than age 35 years in our study population to reduce the inclusion of individuals who might have a reduced fecundity based on age alone. In addition, patients were excluded if they were lost to follow-up, if they had incomplete charts, or if the indication for their IUD insertion was for noncontraceptive reasons. The chart of each of the study participants was reviewed from the time of IUD insertion through March 31, 2013. Clinical data were obtained from office and emergency department visits, phone notes, hospitalizations, laboratory results, and imaging studies.

Demographic data including age at IUD insertion, race, marital status, gravidity, parity, and type of insurance were captured. Information was also obtained regarding the health care provider who initially inserted the IUD. The type of IUD used, either the levonorgestrel intrauterine system or the copper IUD, was identified. Data regarding cervical cultures obtained at the time of IUD insertion were also extracted.

The follow-up time for each patient was defined as the duration of time between IUD insertion and the last visit during which the IUD was documented to be in place by speculum examination, imaging, or reliable history. The primary outcomes evaluated were IUD expulsion, premature discontinuation, and contraceptive failure. Information regarding cervicitis, vaginitis, and pelvic inflammatory disease experienced post-IUD insertion was also collected. Because pelvic inflammatory disease is a clinical diagnosis, we relied on the health care provider's clinical assessment to make the diagnosis. Symptoms such as vaginal bleeding and pelvic pain were not considered side effects of IUD use unless they occurred beyond 2 months after IUD insertion. The severity of the pain was determined after careful review of all documentation in the electronic medical records and it was considered severe if it was described as such by the patient, significantly affected her lifestyle, or required the use of high or long-term doses of analgesics.

For comparison, study participants were divided into three age groups: 13–19, 20–24, and 25–35 years. Because the definition of adolescence is not clearly defined in the literature, these age groups were chosen based on methods from other published national studies that examined birth control methods among women by age group.13–15 Baseline characteristics and outcome measures were presented using means and standard deviations for continuous variables and frequencies and percentages for categorical variables for the overall sample by age categories and IUD type. Bivariate analyses were conducted to examine the differences in baseline characteristics and outcome measures by these groups using analysis of variance and two-sample t tests for continuous variables and χ2 test and Fisher’s exact test for categorical variables.

The relationships between outcome measures and explanatory variables were examined using multivariate logistic regression models to analyze the occurrence of the events during the study and Cox proportional hazards models adjusted for age, race, and IUD type to account for duration to event that varies across individuals. Intercooled Stata 11 was used to conduct all data analyses. A P value of <.05 was used to define statistical significance.

Pairwise differences in patient characteristics and outcomes across age groups should be interpreted with caution because P values adjusted for multiple comparisons were not provided for pairs to conserve space. The study was powered to detect a seven-percentage point difference in expulsion rates (5% in adults compared with 12% adolescents16) at 80% power and a type I error of 0.01 with a two-sided proportions test and a sample size ratio of 5:1 (N1=1,033; N2=237). The achieved sample size with three age groups reached 99% power to detect the trends in the discontinuation rates by age groups (49%, 45%, 37%) observed over the entire study period (two-sided type I error=0.05, Cochrane-Armitage trend test) and would allow us to detect a significant trend (4%, 2%, 1%) in contraceptive failure rates at 84% power.


From all centers combined, we identified a total of 2,523 adolescents and women aged 13–35 years who had an IUD inserted during the study period. Of these females, 328 were lost to follow-up, 31 had an IUD inserted for noncontraceptive reasons, and 26 had incomplete medical records. This resulted in a final study cohort of 2,138 adolescents and women. The overall time of follow-up for this cohort averaged 37±11 months (range 19–57 months).

Demographic data and baseline characteristics stratified by age categories of 13–19, 20–24, and 25–35 years are presented in Table 1. The mean age of our total study population was 25±4.4 years. Forty-five percent of the total study population was African American and 37% was white. Only 13% of our study population was nulliparous, yet 66% of all participants were unmarried. Eighty-two percent of the total study population had the levonorgestrel intrauterine system inserted, and 70% of all IUD insertions were performed by an obstetrician–gynecologist. Adolescents and young women were significantly more likely to be African American, nulliparous, not married, and to have been screened for Neisseria gonorrhea and Chlamydia trachomatis at the time of IUD insertion.

Table 1
Table 1:
Demographics and Baseline Characteristics by Age

Frequency data in regard to the outcomes of interest are shown in Table 2, and the hazard ratios for discontinuation, expulsion, and contraceptive failure are presented in Table 3. The overall rate of IUD expulsion after a mean follow-up of 37 months was 6% and increased from 1% at 1 month of use to 4% at 12 months. No significant differences were found in expulsion rates among the three age groups. However, when IUD expulsion did occur, adolescents and young women between the ages of 13 and 19 years were more likely to have partial IUD expulsion (60%) as compared with the two older cohorts. The two most commonly reported reasons given for IUD expulsion were unknown (81%) and expulsion during menses (10%). After expulsion, 34% of patients elected to have a second IUD placed. The rate of expulsion of the second IUD was 14%.

Table 2-a
Table 2-a:
Frequency of Outcomes With Intrauterine Device Use by Age
Table 2-b
Table 2-b:
Frequency of Outcomes With Intrauterine Device Use by Age
Table 3
Table 3:
Adjusted Hazard Ratios of Risks of Expulsion, Contraceptive Failure, and Discontinuation

The contraceptive failure rates with IUD use were 0.4% at 12 months and 1% after a mean follow-up of 37 months. There were no significant differences among age groups in regard to contraceptive failure, and 100% of the pregnancies identified were intrauterine. Thirty-nine percent of failures occurred after a partial or a complete expulsion not previously identified by the patient or the health care provider. Among those who conceived with an IUD in utero, eight elected removal of the IUD at pregnancy diagnosis, whereas the remaining six females elected to continue the pregnancy with the IUD in utero. The overall rate of spontaneous abortions was 17%; this rate was the same in patients who had their IUD removed as well as the patients who elected to keep their IUD in utero. Of the 23 contraceptive failures, only one preterm delivery was noted, and it was in a patient in whom the IUD was kept in utero.

Intrauterine device discontinuation rates were 19% at 12 months and 41% after a mean follow-up of 37 months. The two most common reasons for discontinuation, irrespective of age and IUD type, were pain (31%) and bleeding (24%). Despite similar rates of IUD discontinuation between age groups at 12 months of use, teenagers and young women younger than the age of 20 years were more likely to request early discontinuation at the end of the total follow-up period.

The rate of cervicitis was inversely related to age with a rate of 13% in the youngest age group. However, no significant difference was noted in pelvic inflammatory disease rates (2%) based on age. Overall, 29% of IUD users reported pain and 30% reported abnormal bleeding. Adolescents and women aged 13–19 years were significantly more likely to report pain. Only three cases of uterine perforation were identified, resulting in an overall rate of 0.3%. They all occurred in parous women, between the ages of 20–24 years, who had their IUD inserted by nurse practitioners.

The frequency of outcomes by IUD type is shown in Table 4. The rate of IUD expulsion was significantly higher in copper IUD users as compared with levonorgestrel intrauterine system users at all time intervals studied (2% compared with 1% at 1 month, 6% compared with 3% at 12 months, and 8% compared with 5% after a mean follow-up of 37 months, respectively). Contraception failure rates were significantly higher in the study participants who had a copper IUD placed as compared with those who had a levonorgestrel intrauterine system placed (1.3% compared with 0.2% at 12 months and 3% compared with 1% in the total follow-up period, respectively). A higher rate of discontinuation was noted in copper IUD users (23%) as compared with levonorgestrel intrauterine system users (18%) at 12 months.

Table 4
Table 4:
Frequency of Outcome by Intrauterine Device Type

Among parous and nulliparous females, no significant differences were seen after a mean follow-up of 37 months in the rates of expulsion (P=.79) contraceptive failure (P=1.0), and premature discontinuation (P=.13). However, nulliparous women were more likely to report pain with IUD use as compared with parous women (P=.02). Furthermore, women who had their IUD inserted by a gynecologist were less likely to discontinue their IUD within the first year of use (P=.02) as compared with those with the IUD inserted by family physicians, nurse practitioners, and physician assistants. Adolescents and women who underwent IUD insertion by a nongynecologist provider did not experience a higher rate of complications such as expulsion (P=.23).


The data from our study demonstrate that, similar to adults, IUD use in adolescents and nulliparous women is effective and associated with low rates of serious complications. This is in contrast to the misconceptions among many women and health care practitioners. Armed with this information, health care providers will be better equipped to educate women, especially teenagers, about the advantages of this method of contraception. The use of an effective long-term contraceptive such as the IUD will help reduce the high rate of unintended pregnancies and should therefore be considered in all females. However, because our data demonstrated that adolescents and young women are more likely to request early discontinuation of their IUDs as a result of pain and irregular bleeding, health care providers must also be aware that this population of patients may require additional counseling in regard to these concerns.

Much of the literature regarding IUD expulsion in adolescents and young women is outdated. In a systematic review published in 2009,16 11 of the 13 articles reviewed were from the 1970s and 1980s and included data primarily on experimental devices or IUDs that are not currently in use.17–20 In these studies, the rate of expulsion reported in the adolescent population was higher than the overall rate of 3–5% and ranged widely from 5% to 22%. Although we found a similar overall rate of expulsion as compared with other studies, 4% at 12 months and 6% after an average follow-up of 37 months, the rate of IUD expulsion in our study was not different based on age or parity. Females should be informed that, although adolescence and nulliparity are not risk factors for expulsion, copper IUD use and a history of prior IUD expulsion both are associated with an increased risk of expulsion in all age groups.

When compared with a recent review of contraception failure by birth control method,21 we found the same reported annual failure rate of 0.2% with the levonorgestrel intrauterine system and a slightly higher rate of 1.3% compared with 0.8% with the copper IUD. Although the copper IUD was noted to have higher failure rates when compared with the levonorgestrel intrauterine system, the failure rates for both types of IUDs were below those of oral contraceptives, patch, ring, condoms, and injectable methods. Only the contraceptive implant had lower method failure rates compared with the IUD.21

Beringer13 recently evaluated the complications and continuation of IUD use in teenagers. This study was a retrospective cohort study of a large sample from across the nation in which data were collected from records of claims paid by an insurance company. Similar to our study, they found that serious complications such as ectopic pregnancy, pelvic inflammatory disease, and perforation are infrequent in all age groups including adolescents and that the levonorgestrel intrauterine system might be a better contraceptive option than the copper IUD as a result of lower rates of complications, discontinuation, and failure.

In our study, half of adolescents and women younger than the age of 20 years (49%) eventually discontinued their IUD within the total study period, although the 12-month continuation rate was less that 25% for all age groups and IUD types. In addition, other studies have shown that when compared with adults, adolescents had higher discontinuation rates for all reversible contraceptive methods with nonlong-acting reversible contraception methods being discontinued at a higher rate than long-acting reversible contraception methods.22 It is possible that adolescents either do not tolerate the side effects of IUDs as well as their older counterparts or that they are not adequately counseled in regard to the type and severity of these adverse effects. As compared with the shorter-acting contraceptives methods, IUDs represent an effective alternative with a low rate of serious complications. Novel interventions are needed to improve adolescent contraceptive continuation in general.

The strengths of our study include the population size, the diversity of our study population, and the standardized collection and interpretation of data among study sites. However, like with similar analyses, our study is limited by its retrospective design. In addition, although we found that patients who were lost to follow-up had similar characteristics to the patients included in the study, they may still have had a higher risk of failure or other adverse outcome that may have altered our results. Also, the group of adolescents using the copper IUD is not large enough for us to draw any definitive conclusions regarding the behavior of the copper IUD compared with levonorgestrel intrauterine system in this age group.


1. Abma JC, Martinez GM, Copen CE. Teenagers in the United States: sexual activity, contraceptive use, and childbearing, national survey of family growth 2006–2008. Vital Health Stat 23 2010:1–47.
2. Finer LB, Zolna MR. Unintended pregnancy in the United States: incidence and disparities, 2006. Contraception 2011;84:478–85.
3. Adolescents and long-acting reversible contraception: implants and intrauterine devices. Committee Opinion No. 539. American College of Obstetricians and Gynecologists. Obstet Gynecol 2012;120:983–8.
4. Benefits and risks of sterilization. Practice Bulletin No. 133. American College of Obstetricians and Gynecologists. Obstet Gynecol 2013;121:392–404.
5. Centers for Disease Control and Prevention (CDC). U.S. Medical Eligibility Criteria for Contraceptive Use, 2010. MMWR Recomm Rep 2010;59:1–86.
6. Yen S, Saah T, Adams Hillard PJ. IUDs and adolescents—an under-utilized opportunity for pregnancy prevention. J Pediatr Adolesc Gynecol 2010;23:123–8.
7. Finer LB, Jerman J, Kavanaugh MC. Changes in use of long-acting contraceptive methods in the United States, 2007–2009. Fertil Steril 2012;98:893–7.
8. Whitaker AK, Johnson LM, Harwood B, Chiappetta L, Creinin MD, Gold MA. Adolescent and young adult women's knowledge of and attitudes toward the intrauterine device. Contraception 2008;78:211–7.
9. Stanwood NL, Bradley KA. Young pregnant women's knowledge of modern intrauterine devices. Obstet Gynecol 2006;108:1417–22.
10. Harper CC, Blum M, de Bocanegra H, Darney PD, Speidel JJ, Policar M, et al.. Challenges in translating evidence to practice: the provision of intrauterine contraception. Obstet Gynecol 2008;111:1359–69.
11. Tyler CP, Whiteman MK, Zapata LB, Curtis KM, Hillis SD, Marchbanks PA. Health care provider attitudes and practices related to intrauterine devices for nulliparous women. Obstet Gynecol 2012;119:762–71.
12. Madden T, Allsworth JE, Hladky KJ, Secura GM, Peipert JF. Intrauterine contraception in Saint Louis: a survey of obstetrician and gynecologists' knowledge and attitudes. Contraception 2010;81:112–6.
13. Berenson AB, Tan A, Hirth JM, Wilkinson GS. Complications and continuation of intrauterine device use among commercially insured teenagers. Obstet Gynecol 2013;121:951–8.
14. Mosher WD, Martinez GM, Chandra A, Abma JC, Willson SJ. Use of contraception and use of family planning services in the US: 1982–2002. Adv Data 2004:1–36.
15. Trussell J, Vaughan B. Contraceptive failure, method-related discontinuation and resumption of use: results from the 1995 National Survey of Family Growth. Fam Plann Perspect 1999;31:64–72.
16. Deans El, Grimes DA. Intrauterine devices for adolescents: a systematic review. Contraception 2009;79:418–23.
17. Lane ME, Sobrero AJ. Experience with intrauterine contraception by adolescent women. Mt Sinai J Med 1975;42:337–44.
18. Jorgensen V. One-year contraceptive follow-up of adolescent patients. Am J Obstet Gynecol 1973;115:484–6.
19. Goldman JA, Dekel A, Reichman J. Immediate postabortion intrauterine contraception in nulliparous adolescents. Isr J Med Sci 1979;15:522–5.
20. Goldman JA, Reichman J. Contraception in the teenager. A comparison of four methods of contraception in adolescent girls. Isr J Med Sci 1980;16:510–3.
21. Trussell J. Contraceptive failure in the United States. Contraception 2011;83:397–404.
22. Rosenstock JR, Peipert JF, Madden T, Zhao Q, Secura GM. Continuation of reversible contraception in teenagers and young women. Obstet Gynecol 2012;120:1298–305.
© 2014 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.