Tyler, Crystal P. PhD; Whiteman, Maura K. PhD; Zapata, Lauren B. PhD; Curtis, Kathryn M. PhD; Hillis, Susan D. PhD; Marchbanks, Polly A. PhD
High unintended pregnancy rates in the United States are thought to result in part from low use of long-acting reversible contraception, such as intrauterine devices (IUDs).1,2 According to 2006–2008 estimates from the National Survey of Family Growth, only 3.4% of United States women of reproductive age reported currently using an IUD for contraception.3 There are two IUDs currently available in the United States, the levonorgestrel-releasing IUD and the copper T IUD. Both IUDs are safe, acceptable, and highly effective, with typical use failure rates similar to those of surgical sterilization.4
When the copper IUD first was introduced to the United States in 1998, it was contraindicated for nulliparous women according to package labeling, primarily because of concerns over subsequent fertility. In 2005, based on newer evidence on the safety of IUDs, the United States Food and Drug Administration approved a label revision that did not restrict use in nulliparous women. Despite American College of Obstetricians and Gynecologists (the College) practice bulletins and recommendations from various organizations, documenting the safety of both the copper IUD and the levonorgestrel-releasing IUD for nulliparous women,2,5–8 previous studies suggest that some health care providers remain concerned about the safety of IUDs for nulliparous women. Specific concerns in nulliparous women include an increased risk of expulsion, pelvic inflammatory disease, and subsequent infertility, along with difficulty and pain with insertion.9–11 Nulliparous women in need of effective contraception represent a significant proportion of the United States population, but the prevalence of long-acting reversible contraception use in this group of women is only 0.5%.12 Health care provider concerns regarding the safety of IUDs and infrequent provision likely play a role in the low prevalence of IUD use among this group of women.9
To offer guidance on the safety of contraceptive methods, including the IUD, the Centers for Disease Control and Prevention adapted global guidance from the World Health Organization and released the United States Medical Eligibility Criteria for Contraceptive Use in June 2010.6 The United States Medical Eligibility Criteria for Contraceptive Use offers the first evidence-based national guidance on the safety of contraceptive methods for women and men with specific characteristics and medical conditions and serves as a source of clinical guidance for health care providers. Based on recommendations from the United States Medical Eligibility Criteria for Contraceptive Use, both the copper IUD and the levonorgestrel-releasing IUD are generally considered safe for nulliparous women.
The purpose of this analysis was to examine the prevalence and predictors of health care provider misconceptions about the safety of IUDs for nulliparous women and of infrequent provision of IUDs to nulliparous women. We sought to extend previous analyses in this area9,10,13 by examining data collected from a nationally representative sample of private and public sector health care providers 5 years after the Food and Drug Administration IUD package revision and before the release of the United States Medical Eligibility Criteria for Contraceptive Use.
MATERIALS AND METHODS
Some methods and definitions for this survey have been described elsewhere.14 Briefly, during December 2009–March 2010, questionnaires were mailed to a random sample of 4,000 health care providers. Based on response rates to similar surveys, we estimated that we would obtain a sufficient sample size for analyses by mailing surveys to 2,000 office-based physicians and 2,000 Title X providers. Office-based physicians were sampled from the American Medical Association Physician Masterfile, which includes information on American Medical Association member and nonmember board-certified physicians residing in the United States and United States territories. We sampled three primary specialties that provide the bulk of family planning services in the United States, including obstetrics and gynecology, family medicine, and adolescent medicine.15 Title X clinics were randomly sampled from a directory maintained by the Office of Population Affairs. Each sampled office-based physician and a randomly selected health care provider from each Title X clinic were eligible to participate if they provided family planning services to women of reproductive age at least twice per week. A family planning service was defined as any service related to postponing or preventing conception and could include a medical examination related to provision of a method, contraceptive counseling, method prescription, or supply visits. A patient could receive a family planning service even if the primary purpose of her visit was not for contraception.
The questionnaire queried health care providers on demographic characteristics, contraceptive method availability at the practice or clinic, attitudes toward the safety of various contraceptive methods for certain groups of women, and the provision of various methods to certain groups of women. Questionnaires were pilot-tested, with physicians representing each targeted specialty, nurse practitioners, certified nurse midwives, and epidemiologists. The initial survey mailing was followed by a reminder postcard and a second survey mailing to nonresponders. Additional efforts to contact nonresponders and to determine eligibility were made by telephone.
Analyses were weighted to correct for nonresponse and for the probability of selection within each office-based physician specialty and for all Title X clinics. Title X data were weighted to reflect the universe of Title X clinics and office-based physician data were weighted proportionate to specialty size to reflect the total membership of board-certified obstetrician–gynecologist (ob-gyn), family medicine, and adolescent medicine physicians.
Of the 2,000 office-based physicians sampled, 478 were excluded because they did not meet the eligibility criteria. Of the 2,000 Title X clinics sampled, 183 were excluded because their providers did not meet the eligibility criteria or the clinic was closed. The Council of American Survey Research Organizations response rate was calculated by assuming that the proportion of health care providers eligible was the same for health care providers with known and unknown eligibility. After accounting for ineligibility, the Council of American Survey Research Organizations response rates were as follows: ob-gyns, 51.8%; family medicine physicians, 44.9%; adolescent medicine physicians, 68.0%; and Title X clinic providers, 77.5%. We restricted this analysis to data obtained from physician and nurse respondents, and thus excluded data from 45 Title X clinic administrators and counselors. Our final analytic sample therefore included 635 office-based physicians and 1,323 Title X clinic health care providers.
Our current analysis examined two main outcomes of interest for each IUD type: health care provider attitudes toward the safety of the IUD for nulliparous women and provision of the IUD to nulliparous women. With respect to safety, respondents were asked to indicate if they considered the method to be very safe, safe, unsafe, very unsafe, or if they were unsure about the safety of the IUD for nulliparous women. Because results were similar when we examined unsafe, very unsafe, and unsure separately, responses in these three categories were grouped and compared with the referent category of respondents who perceived IUDs to be very safe or safe. Because evidence-based guidelines state that IUD use among nulliparous women is safe,2,5,6 we considered responses of unsafe, very unsafe, or unsure to be misconceptions by providers. We examined potential factors related to health care provider misconceptions about the safety of IUDs for nulliparous women, including health care provider characteristics (clinical focus or occupation, sex, years since completion of medical training, receipt of training in interval insertion of IUDs, and the proportion of female patients who receive family planning services), practice or clinic characteristics (region of the United States and practice or clinic availability of the copper IUD and the levonorgestrel-releasing IUD), and patient demographics (proportion of patients with Medicaid or other public assistance, proportion of minority patients, proportion of non–English-speaking patients, and proportion of patients aged younger than 20 years). With respect to clinical focus or occupation (eg, office-based ob-gyn, Title X nurse, and others), midlevel health care providers included certified nurse midwives, nurse practitioners, and physician assistants.
To measure frequency of provision, respondents were asked how often they provided (including by prescription or referral) both the copper IUD and the levonorgestrel-releasing IUD to nulliparous women in the past year: very often, often, rarely, or never. Respondents who indicated that they rarely or never provided IUDs to nulliparous women were considered infrequent health care providers and were compared with the referent category of health care providers who very often or often provided IUDs to nulliparous women. Possible factors related to infrequent provision of IUDs to nulliparous women included health care provider characteristics, practice or clinic characteristics, patient demographics, and misconceptions about the safety of IUDs for nulliparous women. We also examined reasons for infrequent provision of IUDs to nulliparous women. Respondents who indicated that they infrequently provided IUDs to nulliparous women in the past year were additionally asked to indicate reasons why. Potential reasons included concerns about patient safety, contraceptive method unavailability, patient preference for another method, or other. Respondents could select more than one reason.
The Rao-Scott χ2 test statistic, which is used when data are obtained from a complex survey,16 was used to compare characteristics of office-based physicians and Title X clinic health care providers. Using logistic regression, we computed multivariable-adjusted odds ratios (adjusted ORs) and 95% confidence intervals (CIs) to determine factors associated with health care provider misconceptions about the safety of IUDs for nulliparous women and with infrequent provision of IUDs to nulliparous women. Some factors were selected for inclusion in the multivariable model on the basis of previous literature, including clinical focus or occupation, sex, and training in interval IUD insertion; inclusion of other factors was based on significance in univariable models. Effect modification between each of the a priori factors and other variables of interest was assessed using the log likelihood test comparing the full model with the particular interaction term with the reduced model without the interaction term. Office-based physicians and Title X clinic health care providers were combined for multivariable analyses because differences in the two groups were captured by the combined variable “clinical focus or occupation. ” All analyses were conducted with SAS 9.2 using survey methods to account for the complex sample design (SAS Institute). This project was reviewed for human subjects concerns by the Centers for Disease Control and Prevention and was determined to be “nonresearch” or public health practice. Therefore, institutional review board approval was not required.
More office-based physicians than Title X clinic health care providers were male, completed their formal medical training 25 years ago or more, and received training in interval insertion of both IUDs (Table 1). Title X clinic health care providers had a higher proportion of female patients of reproductive age who received family planning services, who paid with Medicaid or other state or federal assistance, who were racial or ethnic minorities, who were non–English-speaking, and who were younger than 20 years of age.
Approximately 30% of respondents had misconceptions about the safety of IUDs for nulliparous women, meaning that they perceived IUDs to be unsafe, very unsafe, or they indicated that they did not know if IUDs were safe for nulliparous women. Among office-based physicians, 16.7% (95% CI 12.7–20.6%) perceived the copper IUD as very unsafe or unsafe, and 13.7% (95% CI 9.8–17.6%) were unsure about the safety of the copper IUDs for nulliparous women (Table 1). Among Title X clinic health care providers, 12.5% (95% CI 11.1–14.0%) perceived the copper IUD as very unsafe or unsafe and 11.9% (95% CI 10.4–13.3%) were unsure about the safety of the copper IUD for nulliparous women. A similar pattern was observed for the levonorgestrel-releasing IUD. Few health care providers indicated that they frequently provided IUDs to nulliparous women in the year before the interview. Office-based physicians (80.2%, 95% CI 76.5–84.0%) were more likely than Title X clinic health care providers (67.5%, 95% CI 65.4–69.6%) to indicate that they infrequently provided copper IUDs to nulliparous women. Provision of the levonorgestrel-releasing IUD did not differ by provider type.
The prevalence of misconceptions about the safety of copper IUDs for nulliparous women varied by clinical focus or occupation, being higher among office-based family medicine physicians than office-based ob-gyns (44.2% compared with 14.7%; adjusted OR 3.20, 95% CI 1.73–5.89; Table 2). A similar pattern was observed for levonorgestrel-releasing IUDs. Among health care providers who were not trained in interval insertion of copper IUDs, the prevalence of misconceptions about safety of these IUDs for nulliparous women was higher than among health care providers who had received training (18.6% compared with 64.1%; adjusted OR 4.72, 95% CI 2.32–9.61). Estimates were similar for the levonorgestrel-releasing IUD. Compared with health care providers for whom IUDs were readily available on-site at their practice or clinic, providers without on-site availability of the copper IUD (adjusted OR 2.18, 95% CI 1.20–3.95) and the levonorgestrel-releasing IUD (adjusted OR 3.45, 95% CI 1.95–6.08) had increased odds of misconceptions about IUD safety. Completion of medical training 25 years ago or more was also associated with health care provider misconceptions about the safety of the copper IUD (adjusted OR 3.69, 95% CI 1.43–9.49).
The prevalence of infrequent provision of copper IUDs and levonorgestrel-releasing IUDs to nulliparous women varied by on-site IUD availability (Table 3). Health care providers without on-site availability of the levonorgestrel-releasing IUD had a higher prevalence of infrequent provision than health care providers with on-site availability (83.7% compared with 57.0%; adjusted OR 2.10, 95% CI 1.22–3.62). A similar pattern was observed for the copper IUD. Compared with health care providers who responded that IUDs are safe for use by nulliparous women, health care providers with misconceptions about the safety of the copper IUD had increased odds of infrequent provision to nulliparous women (adjusted OR 6.04, 95% CI 2.00–18.31). Estimates were similar for the levonorgestrel-releasing IUD. Providing family planning services to less than 75% of female patients of reproductive age was also associated with infrequent IUD provision.
Among health care providers reporting that they rarely or never provided the copper IUD or levonorgestrel-releasing IUD, we examined reasons cited for infrequent provision (Table 4). In general, the most commonly cited reason for infrequent provision of copper IUDs and levonorgestrel-releasing IUDs was patient preference for another method. However, the proportion of health care providers citing patient preference as a reason for infrequent provision varied by health care provider role (P<.05) and was highest among office-based ob-gyns (65.0% for copper IUD, 68.3% for levonorgestrel-releasing IUD) and lowest among Title X nurses (29.4% for copper IUD, 30.2% for levonorgestrel-releasing IUD). Additionally, the proportion of health care providers citing lack of method availability as a reason for infrequent provision varied by health care provider role (P<.05). Lack of availability of the copper IUD and levonorgestrel-releasing IUD was cited most often by office-based adolescent physicians (42.7% for copper IUD, 38.7% for levonorgestrel-releasing IUD), Title X physicians (33.6% for copper IUD, 44.3% for levonorgestrel-releasing IUD), and Title X clinic nurses (29.7% for copper IUD, 35.8% for levonorgestrel-releasing IUD). Concern about patient safety was cited as a reason for infrequent provision by 5–15% of health care providers; this proportion did not vary significantly by health care provider role.
Although most health care providers considered both IUDs safe for nulliparous women, approximately 30% of health care providers in our study perceived IUDs to be very unsafe, unsafe, or were unsure about the safety of IUDs for nulliparous women. Misconceptions about IUD safety were even higher among certain subpopulations, health care providers without on-site IUD availability, and health care providers who were not trained in interval IUD insertion. With respect to provision, the majority of office-based physicians and Title X clinic health care providers rarely or never provided both the copper IUD and the levonorgestrel-releasing IUD to nulliparous women. In adjusted models, IUD availability and misconceptions about safety were most strongly associated with infrequent provision. However, patient preference was most often cited by providers as the reason for infrequently providing IUDs to nulliparous women.
Our results are similar to those of previous studies that indicated that 38–67% of health care providers had misconceptions about the safety of IUDs for nulliparous women.9,10,13,17–20 For example, a 2010 study found that 38% of ob-gyns in the St. Louis area considered the IUD unsafe for nulliparous women,13 with 29% of all providers in the study citing increased pelvic inflammatory disease risk as their main concern. Other studies have examined whether health care providers would recommend IUDs. One study (2010) found that health care providers recommend IUDs for nulliparous women significantly less likely than for parous women with similar gynecologic histories.17 Although our findings were similar to findings from other studies, there were several important differences in study methodology that make our results more generalizable to United States health care providers. Unlike previous studies that were conducted from convenience samples from one city13 or state,10 our sample was nationally representative and included health care providers from all 50 states and United States territories. Unlike an older study,9 ours was conducted after the 2005 Food and Drug Administration liberalization of the copper IUD labeling. Furthermore, our sample included both office-based physician specialists who provide the majority of family planning services and Title X clinic health care providers as opposed to just including health care providers from one clinical focus.9,19,20
Although previous reports have not specifically examined the provision of IUDs to nulliparous women, our results were similar to those of other studies that examined contraceptive provision to the general population of reproductive-aged women. Factors associated with infrequent IUD provision in these studies include seeing fewer contraceptive patients per week13 and a greater number of years since completion of medical training.13 The proportion of patients who are women of reproductive age,18 education and training about contraception,18 and health care provider sector10 have previously been associated with provision of any contraceptive method.
Health care provider attitudes on the safety of IUDs and practices surrounding IUD provision are influenced by a number of factors but may be improved by health care provider training.21 For example, ob-gyns need continuing medical education credits to continue their certification and likely receive more training on the most up-to-date recommendations regarding the safety of various contraceptive methods. Other health care providers such as nurses, who do not necessarily need continuing medical education credits, may not be aware of the latest recommendations related to contraceptive method safety. The College practice bulletins recommend IUDs for nulliparous women and for adolescent patients; however, given recent College guidelines to initiate Pap test screening at age 21 years, patient and health care provider preferences to avoid a pelvic examination may at times lead to prescribing another method of contraception for those younger than age 21 years. Given the influence of patient preference on method provision, efforts are also needed to increase women's knowledge about the benefits and safety of IUDs.
This study has several strengths. We utilized data from a large nationally representative survey of health care personnel who provide the bulk of private, office-based, and publicly funded family planning services. Our sample allowed for the examination of safety attitudes and practices surrounding IUDs for nulliparous women from multiple health care provider types and specialties and provided information after the Food and Drug Administration liberalization of copper IUD packaging. Additionally, data from this nationally representative survey were obtained before the release of the United States Medical Eligibility Criteria for Contraceptive Use. This timeline of data collection allows for the examination of changes in health care provider misconceptions and provision related to dissemination of the United States Medical Eligibility Criteria for Contraceptive Use. The findings of this study are also subject to several limitations. Although our response rate of 47.0% for office-based physicians was similar to that of other surveys targeting physicians, it was fairly low compared with Title X clinic health care providers, and we were unable to determine how respondents and nonrespondents differed with respect to attitudes surrounding the safety of contraceptive methods and frequency of provision. Finally, our survey did not ascertain information on additional factors that could affect attitudes or practices, such as insurance coverage, payment structures for IUDs, or reasons for patient preference for another method, each of which could affect the interpretation of our findings.
These results have important public health implications for the 36 million United States women in need of contraceptive services. Given that the College encourages use of the IUD as a first-line contraceptive choice for nulliparous women,2,5 that IUDs are safe for nulliparous women,6,22 and that IUD continuation and satisfaction rates are high for nulliparous women,23,24 reducing barriers to accessing IUDs is an important first step that may reduce U.S. rates of unintended pregnancy. Given that 30% of surveyed health care providers are not up to date on the latest evidence surrounding the safety of IUDs for nulliparous women, additional training and health care provider education are critical to reduce health care provider biases and to potentially increase provision of IUDs.
1. Tsui AO, McDonald-Mosley R, Burke AE. Family planning and the burden of unintended pregnancies. Epidemiol Rev 2010;32:152–74.
2. Increasing use of contraceptive implants and intrauterine devices to reduce unintended pregnancy. ACOG Committee Opinon No. 450. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:1434–8.
3. Mosher WD, Jones J. Use of contraception in the United States: 1982–2008. Vital Health Stat 23 2010;29:1–44.
4. Hatcher RA, Trussel J, Nelson A, Cates W Jr, Stewart FH, Kowal D. Contraceptive Technology 19th edition. Montvale, NJ: PDR Network; 2008.
5. Intrauterine device. ACOG Practice Bulletin No. 59. American College of Obstetricians and Gynecologists. Obstet Gynecol 2005;105:223–32.
6. Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion, CDC. U S. Medical Eligibility Criteria for Contraceptive Use, 2010. Adapted from the World Health Organization Medical Eligibility Criteria for Contraceptive Use, 4th ed. MMWR Recommend Rep 2010;59(RR-4):1–86.
7. Prager S, Darney PD. The levonorgestrel intrauterine system in nulliparous women. Contraception 2007;75:S12–5.
8. Lyus R, Lohr P, Prager S, Board of the Society of Family Planning. Use of Mirena LNG-IUS and Paragard CuT380A intrauterine devices in nulliparous women. Contraception 2010;81:367–71.
9. Stanwood NL, Garrett JM, Konrad TR. Obstetrician-gynecologists and the intrauterine device: a survey of attitudes and practice. Obstet Gynecol 2002;99:275–80.
10. Harper CC, Blum M, de Bocanegra HT, 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. Hubacher D. Copper intrauterine device use by nulliparous women: review of side effects. Contraception 2007;75:S8–11.
12. Kavanaugh ML, Jerman J, Hubacher D, Kost K, Finer LB. Characteristics of women in the United States who use long-acting reversible contraceptive methods. Obstet Gynecol 2011;117:1349–57.
13. 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.
14. Contraceptive methods available to patients of office-based physicians and title X clinics—United States, 2009–2010. MMWR Morb Mortal Wkly Rep 2011;60:1–4.
15. Scholle SH, Chang JC, Harman J, McNeil M. Trends in women's health services by type of physician seen: data from the 1985 and 1997–98 NAMCS. Womens Health Issues 2002;12:165–77.
16. Rao JNK, Scott AJ. The analysis of categorical data from complex sample surveys: chi-squared tests for goodness-of-fit and independence in two-way tables. J Am Stat Assoc 1981;76:221–230.
17. Dehlendorf C, Ruskin R, Darney P, Vittinghoff E, Grumbach K, Steinauer J. The effect of patient gynecologic history on clinician contraceptive counseling. Contraception 2010;82:281–5.
18. Lohr PA, Schwarz EB, Gladstein JE, Nelson AL. Provision of contraceptive counseling by internal medicine residents. J Womens Health (Larchmt) 2009;18:127–31.
19. Diaz VA, Hughes N, Dickerson LM, Wessell AM, Carek PJ. Clinician knowledge about use of intrauterine devices in adolescents in South Carolina AHEC. Fam Med 2011;43:407–11.
20. Rubin SE, Fletcher J, Stein T, Segall-Gutierrez P, Gold M. Determinants of intrauterine contraception provision among US family physicians: a national survey of knowledge, attitudes and practice. Contraception 2011;83:472–8.
21. Postlethwaite D, Shaber R, Mancuso V, Flores J, Armstrong MA. Intrauterine contraception: evaluation of clinician practice patterns in Kaiser Permanente Northern California. Contraception 2007;75:177–84.
22. Duenas JL, Albert A, Carrasco F. Intrauterine contraception in nulligravid vs parous women. Contraception 1996;53:23–4.
23. Brockmeyer A, Kishen M, Webb A. Experience of IUD/IUS insertions and clinical performance in nulliparous women–a pilot study. Eur J Contracept Reprod Health Care 2008;13:248–54.
24. Wiebe ER, Trouton KJ, Dicus J. Motivation and experience of nulliparous women using intrauterine contraceptive devices. J Obstet Gynaecol Can 2010;32:335–8.