Abstract
Purpose
To identify provider and practice characteristics associated with long-acting reversible contraception (LARC – progesterone contraceptive implants or IUDs [intrauterine devices]) provision among adolescent health care providers.
Methods
We analyzed physician characteristics and self-reported provision of LARC using chi-square analyses. Multivariate logistic regressions identified factors predicting provision of any form of LARC, as well as progesterone contraceptive implants or IUDs specifically.
Results
In logistic regressions, residency training in obstetrics/gynecology or family medicine (rather than internal medicine/pediatrics) was the strongest predictor of LARC provision, particularly for IUDs. Practicing in suburban (rather than urban) and hospital-based (rather than private) settings was associated with lower and higher likelihoods of providing LARC respectively.
Conclusions
Exposure to procedural women’s health training was the strongest predictor LARC provision. Increasing the number of providers offering this type of contraception may have broad reaching consequences for adolescent pregnancy prevention, and may be most easily accomplished via contraceptive implants.
Keywords: adolescent, contraception, long-acting reversible contraception, health care provider, intrauterine device, contraceptive implant
Background
Adolescent pregnancy is associated with an increased risk of suboptimal health outcomes for both mother and child.(1) Despite some success in efforts to reduce the teen pregnancy rate, the United States continues to have one of the highest adolescent pregnancy rates of any developed nation.(1,2) Long-acting reversible contraceptive (LARC) methods such as contraceptive implants and intrauterine devices (IUDs) are the most effective pregnancy prevention options for women of all ages.(3) Studies suggest that much of the recent decline in adolescent pregnancy in the United States may be attributable to improved contraceptive use, especially of more effective methods.(4) The Institute of Medicine has identified expanding young women’s access to these methods as a promising strategy to address their national priority of preventing unintended pregnancy.(5)
Adolescents do not routinely use LARC methods despite their demonstrated effectiveness,(6) and the reasons for this are not well understood. One recent study showed that over 60% of adolescent women choose to contracept using LARC after receiving evidence-based contraceptive counseling and in the absence of financial barriers.(7) This suggests that provider counseling and access barriers may be major factors contributing to the low use of these methods. Lack of clinicians training and comfort in placing these devices for adolescent patients may be another factor perpetuating underutilization.(6)
The goal of our study was to investigate the prevalence of self-reported LARC provision among a group of adolescent health care providers, and to identify provider and practice characteristics that may predict provision of LARC to adolescent patients.
Methods
The data collection methods for this survey have been described in detail elsewhere.(8) For this study, we explored providers’ self-reported provision of LARC methods. We defined LARC providers as those who stated that they personally provided either contraceptive implants or IUDs as part of their routine clinical services. We excluded nurse practitioners (for whom we did not have procedural women’s health training information), physicians whose residency training was in psychiatry or emergency medicine, or providers who denied offering any contraceptive services as part of their regular practice.
We explored LARC provision with respect to provider and practice characteristics using chi-square analyses. Providers were dichotomized into two groups based on their residency training or additional post-residency procedural women’s health training. These groups primarily differentiated between providers whose residency training typically would include procedural women’s health care (obstetrics/gynecology and family medicine – OB/FM) and those whose residency typically would not include such training (internal medicine, pediatrics, and combined medicine-pediatrics – IM/Peds).(9)
We then conducted multivariate logistic regression analyses to identify factors that predicted any LARC provision, as well as provision of contraceptive implants or IUDs specifically. We performed sensitivity analyses to assess whether provision of contraceptive implants could be influenced by potential exposure during training, defined as having finished residency in 2006 or later (after FDA approval of Implanon®).
All data analyses were done with STATA 11.0, and the University of Rochester Research Subjects Review Board approved this study.
Results
Of the 917 U.S. clinician members of SAHM with email addresses, 87% (n=797) received the survey invitation. There were 430 survey participants, resulting in a response rate of 54%. Respondent characteristics are reported in a prior publication,(8) and demographics can be seen in Table 1.
Table 1.
Provider and practice characteristic (chi-square analyses)
| Analytic Sample (n=385) | LARC Providers (n=124) | p-value | |
|---|---|---|---|
| n (%) | n (%) | ||
| Provider Characteristics | |||
| Female | 260 (67.89) | 85 (69.67) | 0.609 |
| Age (years) | 0.407 | ||
| 24–39 | 113 (29.35) | 39 (31.45) | |
| 40–49 | 100 (25.97) | 37 (29.84) | |
| 50–59 | 122 (31.69) | 35 (28.23) | |
| 60 or older | 50 (12.99) | 13 (10.48) | |
| Race | 0.414 | ||
| White | 306 (79.69) | 100 (81.30) | |
| Black | 37 (9.64) | 13 (10.57) | |
| Asian | 27 (7.03) | 5 (4.07) | |
| Other | 14 (3.54) | 5 (4.07) | |
| Latino/Hispanic | 18 (4.71) | 6 (4.96) | 0.877 |
| Residency training | <0.001 | ||
| OB/FM | 45 (11.90) | 35 (28.69) | |
| IM/Peds | 333 (88.10) | 87 (71.31) | |
| Practice Characteristics | |||
| Practice Location | <0.001 | ||
| Urban | 286 (74.48) | 109 (87.20) | |
| Suburban | 78 (20.31) | 11 (8.80) | |
| Rural | 20 (5.21) | 5 (4.00) | |
| Primary Clinical Site | 0.004 | ||
| Private Practice | 54 (14.03) | 6 (4.80) | |
| Academic Medical Center | 125 (32.47) | 42 (33.60) | |
| Hospital-Based Clinic | 135 (35.06) | 50 (40.00) | |
| Other | 71 (18.44) | 27 (21.60) | |
Analytic sample includes providers who met inclusion and exclusion criteria. Cells may not add to overall sample number due to variation in response rates to various questions.
Thirty-two percent of our analytic sample reported providing either form of LARC as a contraceptive option. Among the OB/FM-trained group, 88% reported providing some form of LARC compared to 26% in the IM/Peds group. Forty-seven percent of the OB/FM group reported placing contraceptive implants compared to 24% of the IM/Peds group (data not shown). In logistic regression models (Table 2), presumed exposure to procedural women’s health training was the strongest predictor of LARC provision for both contraceptive implants and IUDs. Practice location and primary clinical site were also associated with LARC provision. Controlling for potential exposure to contraceptive implants during training did not change our results.
Table 2.
Factors associated with LARC provision (logistic regressions)
| Any LARC Method | Contraceptive Implants | IUDs | |
|---|---|---|---|
| OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| Characteristics | |||
| Residency training | |||
| OB/FM (ref: IM/Peds) | 19.40 (5.58–67.37)** | 4.46 (1.72–11.54)* | 83.83 (15.31–458.97)** |
| Practice location | |||
| Suburban (Urban) | 0.20 (0.07–0.57)* | 0.27 (0.10–0.73)* | 0.08 (0.01–0.57)* |
| Rural (Urban) | 0.15 (0.17–1.28) | 0.17 (0.18–1.69) | 0.21 (0.16–2.60) |
p-value <0.05
p-value <0.001
Other variables included in the logistic regression model and not significant include: provider gender, age, race, ethnicity, clinic setting, and adolescent medicine fellowship completion
Discussion
Only a third of our sample reported providing any LARC services, although rates were much higher among providers with procedural women’s health training during or after residency. Exposure to this training was the strongest predictor of any type of LARC provision. However, striking differences emerged between the provision of contraceptive implants and IUDs. OB/FM providers were somewhat more likely to provide contraceptive implants but considerably more likely to provide IUDs than their IM/Peds peers. The skills and equipment necessary to insert IUDs may explain some of this difference, as these can present real barriers to providers who lack procedural gynecologic training, or who practice in settings with few resources.(10) Contraceptive implants, in contrast, require minimal instruction beyond the FDA-mandated manufacturer training, and no special materials beyond the implant inserters themselves.
Limitations of this survey’s findings have been discussed previously,(8) and include lack of generalizability to all providers offering reproductive health services to adolescents. Specific limitations to these analyses include the small sample size of OB/FM-trained clinicians and our use of residency type to reflect exposure to procedural women’s health training.
Conclusions
A third of the providers in our sample reported providing LARC services, and provider residency training appears to be a significant factor in the provision of both contraceptive implants and IUDs. Our results further suggest that exposure to procedural women’s health training is a larger factor for insertion of IUDs than for contraceptive implants.
Reported contraceptive implant placement is lower overall than we might have expected, suggesting an underutilization of implants by adolescent health care providers. As these implants require few procedural skills, it may be easier to increase provider training in contraceptive implants than IUDs. While training will not guarantee an increase in LARC access, it certainly cannot occur without this. Improving access to LARC by increasing the number of providers offering the contraceptive implant would improve LARC access overall and may have broad reaching consequences for adolescent pregnancy prevention.
Acknowledgments
This project was funded by a grant from the Society for Family Planning and supported in part by the Rochester MCH Leadership Education in Adolescent Health T71MC00012-11-01. Kevin Makino is a trainee in the Medical Scientist Training Program funded by NIH T32 GM07356, and was also supported in part by NIH Grant Number UL1 RR024160. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Society for Family Planning, Maternal and Child Health Bureau, National Institute of General Medical Sciences, or NIH. At the time of submission to the Journal of Adolescent Health, this research has also been submitted for consideration as a scientific presentation at the Society for Adolescent Health and Medicine 2013 annual meeting.
Abbreviations
- FDA
US Food and Drug Administration
- AAP
American Academy of Pediatrics
- SAHM
Society for Adolescent Health and Medicine
- LARC
Long Acting Reversible Contraception
- IUD
Intrauterine Device
- OB/FM
obstetrics/gynecology and family medicine
- IM/Peds
internal Medicine, pediatrics, and combined medicine-pediatrics
Footnotes
Conflict of Interest: The authors have nothing to disclose. The first draft of this manuscript was written by Dr. Blumoff Greenberg who did not receive honorarium, grant, or other payment for the production of this manuscript.
Implications and contributions
Increased use of LARC may help mitigate the ongoing problem of unintended adolescent pregnancy. This study implies that limited number of physicians offering LARC may bear some responsibility for adolescents’ underutilization. Formal procedural women’s health training predicted LARC provision, suggesting that provider training may improve adolescents’ access to LARC services.
References
- 1.Klein JD. Adolescent pregnancy: current trends and issues. Pediatrics. 2005 Jul;116(1):281–286. doi: 10.1542/peds.2005-0999. [DOI] [PubMed] [Google Scholar]
- 2.Santelli JS, Melnikas AJ. Teen fertility in transition: recent and historic trends in the United States. Annu Rev Public Health. 2010;31:371–383. 374–383. doi: 10.1146/annurev.publhealth.29.020907.090830. [DOI] [PubMed] [Google Scholar]
- 3.Winner B, Peipert JF, Zhao Q, et al. Effectiveness of long-acting reversible contraception. N Engl J Med. 2012 May 24;366(21):1998–2007. doi: 10.1056/NEJMoa1110855. [DOI] [PubMed] [Google Scholar]
- 4.Santelli JS, Lindberg LD, Finer LB, Singh S. Explaining recent declines in adolescent pregnancy in the United States: the contribution of abstinence and improved contraceptive use. Am J Public Health. 2007 Jan;97(1):150–156. doi: 10.2105/AJPH.2006.089169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.National Academies Press. Initial National Priorities for Comparative Effectiveness Research. Washington DC: National Academies Press; 2009. [Google Scholar]
- 6.Yen S, Saah T, Hillard PJ. IUDs and adolescents--an under-utilized opportunity for pregnancy prevention. J Pediatr Adolesc Gynecol. 2010 Jun;23(3):123–128. doi: 10.1016/j.jpag.2009.09.004. [DOI] [PubMed] [Google Scholar]
- 7.Mestad R, Secura G, Allsworth JE, Madden T, Zhao Q, Peipert JF. Acceptance of long-acting reversible contraceptive methods by adolescent participants in the Contraceptive CHOICE Project. Contraception. 2011 Nov;84(5):493–498. doi: 10.1016/j.contraception.2011.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Coles MS, Makino KK, Phelps R. Knowledge of medication abortion among adolescent medicine providers. J Adolesc Health. 2012 Apr;50(4):383–388. doi: 10.1016/j.jadohealth.2011.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Accreditation Council for Graduate Medical Education (ACGME) [Accessed July 9, 2012];Program Requirements for Residencies in Family Medicine, Obstetrics/Gynecology, Internal Medicine, and Pediatrics. Available at: http://www.acgme.org/acWebsite/home/home.asp.
- 10.Harper CC, Blum M, de Bocanegra HT, et al. Challenges in Translating Evidence to Practice: The Provision of Intrauterine Contraception. Obstet Gynecol. 2008;11(6):1359–1369. doi: 10.1097/AOG.0b013e318173fd83. [DOI] [PubMed] [Google Scholar]