Continuation Rates and Complications of Intrauterine Contraception in Women Diagnosed With Bipolar Disorder

Berenson, Abbey B. MD, MMS; Asem, Humera MD; Tan, Alai PhD; Wilkinson, Gregg S. PhD

Obstetrics & Gynecology:
doi: 10.1097/AOG.0b013e318233beae
Original Research

OBJECTIVE: To estimate continuation rates, complications, and psychiatric hospitalizations among women with bipolar disorder using levonorgestrel-releasing or copper-containing intrauterine devices (IUDs) as compared with those using depot medroxyprogesterone acetate or sterilization for birth control.

METHODS: Data for this cohort study were obtained from a nationwide health insurance claims database on an employed, commercially insured population. Women aged 18–44 years with a prior diagnosis of bipolar disorder (n=849) who were using the levonorgestrel intrauterine system, a copper-containing IUD, depot medroxyprogesterone acetate, or sterilization were evaluated. Outcomes included continuation rates over a 12-month interval, infectious and noninfectious complications, and hospitalizations for bipolar disorder or depression.

RESULTS: Women using an IUD were more likely than those using depot medroxyprogesterone acetate to continue the method for at least 12 months (copper-containing IUD, 86%; levonorgestrel intrauterine system, 87%). In comparison, only 31% of those who initiated depot medroxyprogesterone acetate received three more injections during the next year (P<.001). No significant differences were noted in infectious or noninfectious complications by contraceptive type. Finally, no differences were observed in the number of hospitalizations for bipolar disorder or depression among the four contraceptive groups.

CONCLUSION: More women with bipolar disorder continued using IUDs at one year than women using depot medroxyprogesterone acetate. The rates of complications and psychiatric hospitalizations were not different among women using an IUD, depot medroxyprogesterone acetate, or sterilization.


In Brief

Continuation rates are greater when using intrauterine contraception compared with injectable contraception in women diagnosed with bipolar disorder.

Author Information

From the Departments of Obstetrics and Gynecology and Preventive Medicine and Community Health and the Center for Interdisciplinary Research in Women's Health, University of Texas Medical Branch, Galveston, Texas; and the Department of Obstetrics and Gynecology, University of Illinois, Chicago, Illinois.

Federal support for this study was provided by the Eunice Kennedy Shriver National Institute of Child Health and Human Development as follows: Mid-Career Investigator Award (K24HD04365, A.B.B.); and National Research Service Award institutional training grant for postdoctoral fellows (T32HD055163, H.A.).

Presented at the North American Forum on Family Planning, October 22–24, 2011, Washington, DC.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health and Human Development nor the National Institutes of Health.

Corresponding author: Abbey B. Berenson, MD, MMS, UTMB, 301 University Boulevard, Galveston, TX 77555-0587; e-mail:

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

Article Outline

Bipolar disorder spectrum occurs in 4.4%, or nine million, of women in the United States according to the first lifetime prevalence estimates from the 2007 National Comorbidity Survey Replication.1 Age of onset ranges from 18 to 22 years with women of childbearing age demonstrating the highest prevalence during a 12-month period: ages 18–29 years, 4.7%; ages 30–44 years, 3.5%).2,3 Women with bipolar disorder are at high risk for unintended pregnancies as a result of multiple issues including sexual risk behavior associated with impulsivity of mania, not using contraception, or using ineffective contraception.46 They are also more likely to have had more than one partner7,8 and to engage in sex trading compared with the general U.S. population.4 Consistent with these high-risk sexual behaviors, a survey of 136 women with bipolar disorder reported that only 58.8% were using contraception. The most common methods were oral contraceptives (16.2%) and tubal ligation (16.3%). Other methods included condoms (12.8%), intrauterine contraception (4.4%), and partner vasectomy (2.8%).5

Recommending appropriate contraceptives for patients with bipolar disorder is challenging because of certain issues related to this population. First, many women with bipolar disorder may be treated with mood stabilizers, including antiepileptics, which can affect drug levels of oral contraceptives, thus impairing contraceptive efficacy.912 In addition, some evidence indicates that oral contraceptives may decrease levels of certain antieplileptics.11,12 The second issue facing this population is noncompliance, which would indicate long-acting reversible contraceptives, such as an injectable depot medroxyprogesterone acetate and intrauterine devices (IUDs) to be more desirable options than oral or barrier methods. However, because of high-risk sexual behavior-associated impulsivity, women with bipolar disorder may be at higher risk for sexually transmitted infections.13 As a result, some clinicians may not recommend IUDs to this population.

Another challenge in providing contraceptives to patients with bipolar disorder is the lack of published studies that evaluate complications of different methods in this population. The purpose of this study was to estimate complications in women with bipolar disorder who used an IUD compared with women who used depot medroxyprogesterone acetate or sterilization. Our hypothesis was that women using IUDs would have similar rates of complications compared with women using depot medroxyprogesterone acetate or sterilization. Specifically, we used medical claims data to investigate complications and psychiatric hospitalizations among women using levonorgestrel-releasing and copper IUDs compared with women using depot medroxyprogesterone acetate or sterilization.

Back to Top | Article Outline


Data for this cohort study were obtained from a nationwide U.S. health claims database consisting of electronic health insurance claims from an employed, commercially insured population with dependents. A number of validation studies has documented use of these data for studies of health care use and outcomes.1416 Medical claims from 2001 to 2006 were evaluated. All patient identifiers were removed before data being made available for research purposes. The University of Texas Medical Branch institutional review board granted an exemption for this protocol.

To examine the effects of using long-acting reversible contraceptives among women with bipolar disorder, we first identified all women aged 18–44 years who received one of four different highly effective contraceptive methods: intrauterine contraception, copper-containing IUDs, or a levonorgestrel intrauterine system; depot medroxyprogesterone acetate injections; or permanent sterilization between 2001 and 2005 and had 12 months of continuous follow-up. All depot medroxyprogesterone acetate users had received four injections within 12 months, which is the recommended standard of care (Fig. 1). A total of 184,640 women were identified, of whom 849 were coded as having a diagnosis of bipolar disorder (bipolar disorder I, bipolar disorder II, or subthreshold bipolar disorder) during the year before use of long-acting reversible contraceptives. Eight of these women were excluded because they were immunocompromised and thus at increased risk of infection. Of the remaining 841 women diagnosed with bipolar disorder, 407 had undergone sterilization, 113 had used a copper-containing IUD, 139 had used the levonorgestrel intrauterine system, and 182 had received four injections of depot medroxyprogesterone acetate within 12 months. For women in each of the four contraceptive groups, age and the region of residence in the United States were described.

Outcomes examined for each of the four contraceptive groups included rates of contraceptive continuation over a 12-month interval, infectious and noninfectious complications, and rates of hospitalization for bipolar disorder. Because 40% of bipolar patients, particularly women, are diagnosed initially with unipolar depression, we also included hospitalizations for depression.17,18 Efficacy of the four methods has been well established and was not examined in this study. In instances in which an IUD was discontinued within the first year of use, these complications were evaluated only for the duration of use. Infectious outcomes included gonococcal infections, inflammatory diseases of the female reproductive tract, and cervicitis or endocervicitis. Noninfectious complications included pain associated with female genital organs (dyspareunia, dysmenorrhea, or premenstrual tension syndrome), disorders of menstruation (absences of menstruation, scanty or infrequent menstruation, excessive or frequent menstruation, irregular menstrual cycles, metrorrhagia, postcoital bleeding, dysfunctional or functional uterine hemorrhage, or unspecified), accidental puncture or laceration, and mechanical complications of IUDs. As a result of the vague nature for the last code, all claims within 1 year of this diagnosis were reviewed. In addition, the ectopic pregnancy rate was examined for each group. Chi-square tests were used to compare categorical variables and one-way analysis of variance to compare continuous variables.19 The exact Pearson's chi-square test was used to compare the rates of hospitalizations for bipolar disorder and depression among the contraceptive groups.

Back to Top | Article Outline


Sterilization (48.4%) was the most common method of long-acting birth control selected by women diagnosed with bipolar disorder followed by depot medroxyprogesterone acetate (21.6%), levonorgestrel intrauterine system (16.5%), and a copper-containing IUD (13.4%). Women undergoing sterilization were significantly older than those using the three reversible contraceptive methods examined. No significant differences were observed in the type of birth control used across geographic regions among these women with bipolar disorder (Table 1).

When the three long-acting reversible contraceptive methods (a copper-containing IUD, levonorgestrel intrauterine system, and depot medroxyprogesterone acetate) were compared, a significant difference was observed in continuation rates. (Table 2). Among women who received a copper-containing IUD or levonorgestrel intrauterine system, 86% and 87%, respectively, were still using the method 12 months later. In comparison, only 31% of those who received a depot medroxyprogesterone acetate injection during the study interval received three more injections during the next year (P<.001). With regard to the effect of these contraceptive methods on bipolar disorder, no significant differences were observed in the number of hospitalizations for bipolar disorder or depression among the four contraceptive groups (Table 3).

Overall, we did not observe any differences by contraceptive type in infectious or noninfectious complications (Table 2). Cervicitis was the most common infectious complication across all contraceptive groups (1.8% copper-containing IUD, 3.6% levonorgestrel intrauterine system, 1.1% depot medroxyprogesterone acetate, and 3.0% sterilization), whereas dysmenorrhea was the most common noninfectious complication (5.3% copper-containing IUD, 2.9% levonorgestrel intrauterine system, 2.8% depot medroxyprogesterone acetate, and 4.9% sterilization). Three IUD users (two copper-containing IUDs and one levonorgestrel intrauterine system) were coded for “mechanical complication of an IUD.” As a result of the vague description of this code, the trajectories for these three cases were reviewed in detail. One woman with a copper-containing IUD was coded with this diagnosis 2 days after insertion followed by an emergency room encounter for pain associated with female genital organs 4 days later. Four days after the emergency room encounter, the copper-containing IUD was removed. This sequence of events suggests that the mechanical complication may have been a partial expulsion. The other two patients did not have any testing or procedures related to their mechanical complication, so the exact nature of the complication could not be determined.

Furthermore, no significant difference was observed in the number of ectopic pregnancies that occurred in each group (P=.14). One patient was coded with this diagnosis 2 weeks after her levonorgestrel intrauterine system insertion, whereas another had an ectopic pregnancy 7 months after insertion of a copper-containing IUD. Both underwent a laparoscopic procedure to treat the ectopic pregnancy. No patients in the sterilization or depot medroxyprogesterone acetate groups had a claim for the diagnosis or treatment of an ectopic pregnancy during the 12 months of follow-up.

Back to Top | Article Outline


This study examined complications associated with sterilization and three types of long-acting reversible contraceptives among women previously diagnosed with bipolar disorder. We observed that rates of infectious and noninfectious complications were low among this population of women using sterilization and different types of long-acting reversible contraceptive. Furthermore, hospitalizations resulting from bipolar disorder or depression did not differ between women using different types of contraception.

Similar to the general population, IUD continuation rates were higher than depot medroxyprogesterone acetate after 1 year of use. In fact, over 80% of women who selected either a levonorgestrel intrauterine system or copper-containing IUD continued their chosen method for at least 12 months. This is in agreement with other studies on women without a history of mental illness.10,20 In contrast, only 31% of women who began depot medroxyprogesterone acetate received four injections in a 12-month period. This continuation rate is much lower than reported rates that demonstrate approximately 50% of women who initiate depot medroxyprogesterone acetate continue this method for at least 1 year.10,2125

When we re-examined results of 14 international clinical studies conducted by the World Health Organization, we found that although the average continuation rate at 1 year was 51% for depot medroxyprogesterone acetate, the rate ranged from 25% to 67% on individual investigations.24 Similarly, some studies in U.S. women have reported 12-month continuation rates on depot medroxyprogesterone acetate ranging from 21.3% to 28.6%.2628 In line with our findings, another study generated from a different U.S. insurance claims database indicates that nearly 60% of new depot medroxyprogesterone acetate users did not receive a second injection at 3 months.29,30 Perhaps differences in study settings such as international compared with U.S. populations and formal clinical research studies compared with actual use as documented in medical claims data have contributed to the wide range of continuation rates reported for depot medroxyprogesterone acetate. Also, a copayment was required of the women reflected in the medical claims database. In addition, they would not have received financial incentives nor extended supportive counseling that are usually provided to research participants in depot medroxyprogesterone acetate studies.

Use of a large claims database allowed us to examine usage of four different types of contraception and related complications among reproductive-aged women diagnosed with bipolar disorder, thus providing new insights into the importance of choosing long-acting reversible contraceptives for these women. However, there are several limitations to this study that could have affected our conclusions. First, the population consisted of women with commercial health insurance, so our findings may not be applicable to those who do not have this level of coverage. Also, claims data are generated from past patient encounters, precluding randomization of women to different methods of contraception. Because we detected few complications, we may not have had the statistical power to detect differences. In addition, women may not seek medical help for minor side effects associated with contraceptive use. Thus, some complications are likely underreported in these data. However, we would anticipate that major complications were reported and coded. Medical charts were not available for review, so we were not able to determine reasons for contraceptive discontinuation. Future studies should address these issues to help provide more information in addressing the contraceptive needs of women with bipolar disorder.

Despite these limitations, we have demonstrated that women with bipolar disorder who desire a reversible contraceptive method are appropriate candidates for long-acting reversible contraceptives. Furthermore, women with this disorder are more likely to continue using IUDs than depot medroxyprogesterone acetate for at least 1 year for contraception. Although findings from claims data possess certain limitations, mined information can provide a valuable snapshot of actual treatment practices and outcomes within a defined population. Bipolar disorder and associated medication challenges are sometimes poorly understood by obstetrician–gynecologists. Likewise, the contraceptive needs of women with the disorder may not be adequately addressed by those in psychiatry. Our study provides relevant information for improved clinical decision-making by obstetrician–gynecologists whose patients include women with bipolar disorder.

Back to Top | Article Outline


1. Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RM, Petukhova M, et al.. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry 2007;64:543–52.
2. Kessler RC, Merikangas KR, Wang PS. Prevalence, comorbidity, and service utilization for mood disorders in the United States at the beginning of the twenty-first century. Annu Rev Clin Psychol 2007;3:137–58.
3. National Comorbidity Survey Replication (NCS-R). 12-month prevalence estimates 2007. Table 2. Available at: Retrieved May 24, 2011.
4. Meade CS, Graff FS, Griffin ML, Weiss RD. HIV risk behavior among patients with co-occurring bipolar and substance use disorders: associations with mania and drug abuse. Drug Alcohol Depend 2008;92:296–300.
5. Magalhães PV, Kapczinski F, Kauer-Sant'Anna M. Use of contraceptive methods among women treated for bipolar disorder. Arch Womens Ment Health 2009;12:183–5.
6. Curtis V. Women are not the same as men: specific clinical issues for female patients with bipolar disorder. Bipolar Disord 2005;7(suppl 1):16–24.
7. Coverdale JH, McCullough LB, Chervenak FA. Sexually transmitted diseases and unwanted pregnancies in chronically ill psychiatric patients. Med J Aust 1997;166:231–2.
8. Coverdale JH, Turbott SH, Roberts H. Family planning needs and STD risk behaviours of female psychiatric out-patients. Br J Psychiatry 1997;171:69–72.
9. Crawford P. Interactions between antiepileptic drugs and hormonal contraception. CNS Drugs 2002;16:263–72.
10. Centers for Disease Control and Prevention (CDC). US medical eligibility criteria for contraceptive use, 2010. MMWR Recomm Rep 2010;59:1–86.
11. Dutton C, Foldvary-Schaefer N. Contraception in women with epilepsy: pharmacokinetic interactions, contraceptive options, and management. Int Rev Neurobiol 2008;83:113–34.
12. Gaffield ME, Culwell KR, Lee CR. The use of hormonal contraception among women taking anticonvulsant therapy. Contraception 2011;83:16–29.
13. Carey MP, Carey KB, Maisto SA, Schroder KE, Vanable PA, Gordon CM. HIV risk behavior among psychiatric outpatients: association with psychiatric disorder, substance use disorder, and gender. J Nerv Ment Dis 2004;192:289–96.
14. Lanza LL, Dreyer NA, Schultz NJ, Walker AM. Use of insurance claims in epidemiologic research: identification of peptic ulcers, GI bleeding, pancreatitis, hepatitis, and renal disease. Pharmacoepidemiol Drug Saf 1995;4:239–48.
15. Enger C, Cali C, Walker AM. Serious ventricular arrhythmias among users of cisapride and other QT-prolonging agents in the United States. Pharmacoepidemiol Drug Saf 2002;11:477–86.
16. Johannes CB, Koro CE, Quin SG, Cutone JA, Seeger JD. The risk of coronary heart disease in type 2 diabetic patients exposed to thiazolidinediones compared to metformin and sulfonylurea therapy. Pharmacoepidemiol Drug Saf 2007;16:504–12.
17. McElroy SL. Bipolar disorders: special diagnostic and treatment considerations in women. CNS Spectr 2004; 9(suppl 7):5–18.
18. Saunders KEA, Goodwin GM. The course of bipolar disorder. Advances in Psychiatric Treatment 2010;16:318–28.
19. Agresti A. An introduction to categorical data analysis, 2nd ed. New York (NY): John Wiley & Sons; 2007.
20. Peipert JF, Zhao Q, Allsworth JE, Petrosky E, Madden T, Eisenberg D, et al.. Continuation and satisfaction of reversible contraception. Obstet Gynecol 2011;117:1105–13.
21. Paul C, Skegg DC, Williams S. Depot medroxyprogesterone acetate. Patterns of use and reasons for discontinuation. Contraception 1997;56:209–14.
22. Hubacher D, Goco N, Gonzalez B, Taylor D. Factors affecting continuation rates of DMPA. Contraception 1999;60:345–51.
23. Berenson AB, Odom SD, Breitkopf CR, Rahman M. Physiologic and psychologic symptoms associated with use of injectable contraception and 20 microg oral contraceptive pills. Am J Obstet Gynecol 2008;199:351.e1–12.
24. Vaughan B, Trussell J, Kost K, Singh S, Jones R. Discontinuation and resumption of contraceptive use: results from the 2002 National Survey of Family Growth. Contraception 2008;78:271–83.
25. Draper BH, Morroni C, Hoffman M, Smit J, Beksinska M, Hapgood J, et al.. Depot medroxyprogesterone versus norethisterone oenanthate for long-acting progestogenic contraception. The Cochrane Database of Systematic Reviews 2006, Issue 3. Art. No.: CD005214. DOI: 10.1002/14651858.CD005214.pub2.
26. Westfall JM, Main DS, Barnard L. Continuation rates among injectable contraceptive users. Fam Plann Perspect 1996;28:275–7.
27. Sangi-Haghpeykar H, Poindexter AN 3rd, Bateman L, Ditmore JR. Experiences of injectable contraceptive users in an urban setting. Obstet Gynecol 1996;88:227–33.
28. Raine TR, Foster-Rosales A, Upadhyay UD, Boyer CB, Brown BA, Sokoloff A, et al.. One-year contraceptive continuation and pregnancy in adolescent girls and women initiating hormonal contraceptives. Obstet Gynecol 2011;117:363–71.
29. Halpern V, Lopez LM, Grimes DA, Gallo MF. Strategies to improve adherence and acceptability of hormonal methods of contraception. The Cochrane Database of Systematic Reviews 2011, Issue 4. Art. No.: CD004317. DOI: 10.1002/14651858.CD004317.pub3.
30. Murphy PA, Brixner D. Hormonal contraceptive discontinuation patterns according to formulation: investigation of associations in an administrative claims database. Contraception 2008;77:257–63.
© 2011 by The American College of Obstetricians and Gynecologists.