Contraception xx (2014) xxx – xxx
Original research article
Effect of an educational script on postpartum contraceptive use: a randomized controlled trial☆,☆☆,★ Jennifer H. Tang a,⁎, Rosalie C. Dominik b , Matthew L. Zerden a , Sarah B. Verbiest a , Seth C. Brody a, c , Gretchen S. Stuart a a
Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, NC 27523, USA b Department of Biostatistics, University of North Carolina, Chapel Hill, NC 27523, USA c Department of Obstetrics and Gynecology, WakeMed Health and Hospitals, Raleigh, NC, USA Received 31 October 2013; revised 9 February 2014; accepted 23 March 2014
Abstract Objectives: Short interpregnancy intervals are associated with adverse perinatal outcomes. Increased postpartum use of long-acting reversible contraception (LARC) could reduce short interpregnancy intervals. Therefore, the primary objective of our study was to evaluate if a postpartum educational script about LARC (LARC script) could increase postpartum LARC utilization at the 6-week postpartum visit. Study design: We completed a randomized controlled trial (1:1) of 800 postpartum women in Raleigh, North Carolina. Women were recruited from the postpartum unit and randomized to receive (intervention) or not receive (control) the LARC script prior to hospital discharge. We conducted follow-up phone interviews to assess LARC use after their 6-week postpartum visit. We used Pearson's chi-squared test to compare LARC use between arms. Results: Between May 2011 and January 2012, 400 women were randomized to each arm. Three hundred sixty-nine women (92.3%) in each arm were successfully contacted after their 6-week postpartum visit. LARC use was reported by 17.6% and 13.3% of women in the intervention and control arms, respectively (p=.103). Conclusions: The LARC script did not increase LARC utilization at the 6-week postpartum visit. Future studies should evaluate interventions that address both educational and systematic barriers to postpartum LARC uptake. Implications: Use of a postpartum educational script focused on the intrauterine device and contraceptive implant did not increase their utilization after the 6-week postpartum visit. Future studies should evaluate interventions that address both educational and systematic barriers to postpartum contraceptive uptake. © 2014 Elsevier Inc. All rights reserved. Keywords: Contraception; Reversible; Patient education; Intrauterine device; Contraceptive implant
1. Introduction Short interpregnancy intervals are associated with adverse perinatal outcomes [1]. In the U.S. and North Carolina, 38% and 42% of pregnancies have interpregnancy intervals less
☆ Funding Acknowledgement: Funded by the Fellowship in Family Planning Grant #3631. ☆☆ Conflicts of Interest: The authors report no conflict of interest. ★ Clinical Trial Registration: ClinicalTrials.gov, www.clinicaltrials. gov, NCT01443533. ⁎ Corresponding author. 3031 Old Clinic Building, CB #7570, Chapel Hill, NC 27599-7570, USA. Tel.: +1-919-843-9546; fax: +1-919-843-9437. E-mail address: [email protected] (J.H. Tang).
http://dx.doi.org/10.1016/j.contraception.2014.03.017 0010-7824/© 2014 Elsevier Inc. All rights reserved.
than 18 months, respectively [2,3]. Postpartum contraceptive counseling may play a role in decreasing short interpregnancy intervals. A Cochrane review of eight randomized controlled trials found that postpartum contraceptive counseling can increase postpartum contraceptive use [4]. However, none of the trials focused on the most effective reversible contraceptives, long-acting reversible contraception (LARC) [5,6]. In the U.S., four LARC methods are available: a hormonal implant, a copper intrauterine device (IUD) and two levonorgestrel-releasing intrauterine systems. The American College of Obstetricians and Gynecologists has recommended increased LARC use to reduce the high rate of unintended pregnancy [6]. Increased postpartum use of LARC could also prevent short interpregnancy intervals
2
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
since they all provide effective contraception for at least 3 years [7,8]. Three main barriers prevent increased use of LARC: low patient awareness, high upfront costs and lack of health care provider knowledge or skills [5,9,10]. The Contraceptive CHOICE project overcame these barriers and achieved a LARC uptake rate of 77% in 7486 women by using three interventions: (1) a brief scripted introduction to LARC methods (LARC script), (2) removal of financial barriers in obtaining LARC and (3) ensuring that all eligible women interested in the IUD had a provider to place it [11,12]. These interventions could also increase LARC use in the postpartum population. However, the latter two interventions may not be easily replicable in other settings. The LARC script could be easily modified for the postpartum population. Therefore, the primary objective of this study was to determine if receiving a modified version of the LARC script in the immediate postpartum period would increase postpartum LARC use. We hypothesized that women who received our LARC script during their postpartum hospitalization would be more likely to report LARC use after their 6-week postpartum visit, compared to women who did not receive the script. Secondary objectives were to assess if the script would increase overall contraceptive use or interest in using a LARC method after the 6-week postpartum visit.
2. Materials and methods This study was a parallel (1:1) randomized controlled trial of 800 women admitted to the postpartum unit of WakeMed Hospital between May 2011 and January 2012. The study protocol followed CONSORT guidelines and was approved by the Institutional Review Boards of the University of North Carolina and WakeMed Hospital [13]. WakeMed Hospital is a large private nonprofit hospital in Raleigh, North Carolina, with about 5000 deliveries annually. Immediate postpartum insertion of LARC was not available there during the study period, and no standard procedures for postpartum contraceptive counseling existed. The research assistants looked through the delivery log every weekday morning and recruited women who were likely to be discharged first. Due to involvement in other studies, they could not recruit consecutively every day and instead recruited until they reached a minimum quota of 25 randomized women per week. The eligibility criteria for the study were as follows: (1) admission to the postpartum unit, (2) delivery of a live infant N 24 weeks gestational age, (3) age 14–45 years, (4) ability to speak English or Spanish fluently, (5) willing to be contacted by phone until at least 8 weeks after delivery, (6) no history of a tubal ligation or hysterectomy, (7) current partner has not had a vasectomy, (8) no use of fertility treatment to conceive this pregnancy and (9) no previous randomization into the study.
The research assistants approached eligible women in their postpartum hospital rooms. Women were told that the study's purpose was to learn about contraceptive use among postpartum women and they would be randomized to receive or not receive an experimental educational message. Women underwent informed consent if they agreed to participate and completed a 20-question Baseline Survey via face-to-face interviews. Additional demographic and obstetric history data were abstracted from their medical records. After each interview, the research assistant opened a sequentially numbered opaque sealed envelope. The envelope contained a piece of paper that was either blank (control arm) or had the 1-min LARC script (Appendix A) in English and Spanish (intervention arm). Participants allocated to the intervention arm had the LARC script read aloud to them and were handed the script. If the participant wanted more information about a LARC method, she was given company promotional information approved by the Food and Drug Administration for that method. The research assistants helped the participants schedule their 6-week postpartum visits (the standard of care at the hospital) as needed. Data were then collected from two follow-up phone surveys. Follow-Up Survey #1 occurred 1–2 weeks postpartum and consisted of a 1-min questionnaire to confirm the participant's 6-week postpartum visit date. Phone calls to complete Follow-Up Survey #2 began the day after the participants' scheduled 6-week postpartum visit and involved a 10-min survey with 25 questions. Follow-Up Survey #2 included questions about whether or not the participant was using LARC (primary outcome), whether or not she was interested in using LARC (secondary outcome) and which contraceptive method she was using (secondary outcome). To assess if a participant was interested in using LARC, we asked, “Are you interested in using the intrauterine device (the IUD) or the birth control implant (Implanon) for birth control?” If the participant answered “Yes”, she was then asked “Which one are you interested in using?” At the end of Follow-Up Survey #2, all participants were given a chance to learn more about LARC methods by having the LARC script read to them. A multi-tiered contact system was used to track the subjects. If the participant could not be reached to complete Follow-Up Survey #2 after five attempts to her primary or secondary phone numbers, she was emailed and/or paged. If there still was no response, the participant's contact persons, postpartum clinic and pediatric clinic were contacted if she had given permission to contact them during enrollment. Finally, if all previous attempts failed and the participant was still within 3 months of her scheduled 6-week postpartum visit, the survey was mailed to the patient, with a prestamped and pre-addressed return envelope. The participant was considered lost-to-follow-up if Follow-Up Survey #2 was not returned within 1 month of being sent. Simple 1:1 randomization was achieved via computergenerated randomly permuted block randomization, with block sizes of 6, 8 and 10. The randomization scheme was
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
developed and stored by an independent statistician at the North Carolina Translational and Clinical Sciences Institute and used to prepare the allocation envelopes. The study investigators were blinded to the randomization until data analysis. The potential for bias was decreased by designating that the research assistant who opened the allocation envelope for a participant was not the same research assistant who completed that participant's Follow-Up Surveys. For our sample size calculations, we assumed that 20% of women in the control arm would be using LARC after their 6-week postpartum visit, based on LARC use rates reported among women 2–3 months postpartum in the North Carolina Pregnancy Risk Assessment Monitoring Survey [3]. An analysis population with at least 626 women (313 in each arm) was required to provide 80% power (using a two-sided alpha of 0.05) to detect an absolute 10% increase to 30% in LARC use in the intervention arm [14]. Anticipating a maximum drop-out rate of 20% at the time of Follow-Up Survey #2, we planned to randomize 800 participants. Descriptive statistics were used to summarize the baseline variables by assigned arm. For analysis of LARC use and any contraceptive use after the 6-week postpartum visit, we used Pearson's chi-square test to compare the proportions of use between the two arms and calculated the risk ratio (RR) and their 95% confidence intervals (CIs). Interest in a LARC method was considered an ordinal variable with three categories: using a LARC method, interested but not using a LARC method and not interested and not using a LARC method. The row mean score chi-square test was used to compare the distributions of the ordinal LARC interest variable between study arms [15]. During data analysis, we found that 34 participants were not allocated to their randomly assigned arm due to error in preparing the allocation envelopes. Therefore, we first performed all planned analyses using the arm assigned by the computergenerated randomization. We then performed sensitivity analyses using the actual arm allocated to the participants. We repeated the LARC use comparison by mode of delivery and used modified Poisson regression to assess interaction between mode of delivery and intervention [16]. Based on prior studies, our hypothesis was that women who had cesarean deliveries would desire longer birth intervals than women who delivered vaginally [17,18]. Differences between the two randomized arms were tested at the 0.05 alpha level, whereas testing for the interaction between mode of delivery and the randomized arm was performed at the 0.10 alpha level. All analyses were done using Stata 12.0 (StataCorp LP, College Station, TX). Study data were managed using the Web-based REDCap (Research Electronic Data Capture) electronic data system [19].
3. Results The study was conducted from May 2011 through April 2012. Nine hundred eighty-nine women were screened, 800
3
(80.9%) enrolled and 738 (92.3%) of the enrolled women completed the study (Fig. 1). The computer-generated randomization sequence assigned 400 women to receive the LARC script (intervention arm) and 400 women to receive no LARC script (control arm). Due to an error in preparing the allocation envelopes, 394 women were actually allocated to the LARC script (15 incorrectly), and 403 women were allocated to no LARC script (19 incorrectly). Participants in both arms had similar baseline characteristics, except that women in the intervention arm were more likely to desire no future children (47.8%) than in the control arm (38.2%) (Table 1). Only a third were trying for pregnancy at the time of conception, and about 80% did not want another pregnancy within 2 years. Approximately 10% had previously used LARC, and almost 40% intended to use LARC as postpartum contraception. Of the 738 women who completed Follow-Up Survey #2, 620 (84.0%) reported attending a postpartum visit. The median time between the 6-week postpartum visit and Follow-up Survey #2 was 11 days, with a range 0–121 days. In the intervention arm, 17.6% reported using a LARC method after their 6-week postpartum visit, compared to 13.3% in the control arm (RR 1.3; 95% CI 0.9–1.9; p=.10) (Table 2). Results of the sensitivity analysis using actual allocation were similar, with no significant difference between the two arms. We found a significant difference in LARC interest between the two arms (Table 2). Thirty one percent of women in the intervention arm and 26% in the control arm were interested in but not yet using LARC. The comparison of the ordinal interest variable between the two arms showed significantly increased interest in the intervention arm compared to the control arm (p=.01). Sensitivity analysis using actual allocation revealed similar results. Of note, in the intervention arm, 49% of participants were either using or interested in using LARC, compared to 39% in the control arm. No statistically significant difference was found between arms for use of any contraceptive method (RR 1.0; 95% CI 1.0–1.1; p=.32) (Table 2). The interaction between mode of delivery and script was also not statistically significant (p=.28).
4. Discussion In our study, administration of a LARC script to postpartum women did not significantly increase postpartum LARC use. However, the script significantly increased interest in using LARC. The difference between the effect of the script on LARC interest and LARC use may suggest that many participants interested in using LARC were unable to obtain it. Prior to randomization in the study, almost 40% of participants planned to use LARC for postpartum contraception, but less than half were actually using it after their 6-week postpartum visit. Barriers to receiving LARC, such
4
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
Fig. 1. Study flow chart.
as delayed timing of the postpartum visit, likely exist and should be further investigated. Four previous randomized controlled trials have evaluated the impact of immediate postpartum contraceptive education on postpartum contraceptive use. Two studies found that the educational intervention significantly increased overall postpartum contraceptive use, one of which also led to a 10% increase in postpartum IUD use [20,21]. The educational interventions in these studies were each 20-min long, but neither was focused on LARC nor was they performed in the U.S. The other two studies were performed in the U.S. but were inconclusive [22,23]. The first study was underpowered to show a difference in contraceptive use, and the second only assessed satisfaction with the educational intervention. Differences in study populations and types of interventions may account for the contrasting results with our study, which found that neither postpartum IUD nor overall contraceptive use increased with our educational intervention. If future studies in the U.S. are
performed, they should consider using longer educational counseling sessions. Our study's strengths include its large size, blinded follow-up and randomization of the study intervention. Although we discovered an error in our randomization assignment, sensitivity analyses confirmed minimum impact on the primary and secondary outcomes. A potential weakness is that we did not complete Follow-Up Survey #2 at the same postpartum interval for all participants. Some participants had their 6-week postpartum visits scheduled as late as 10 weeks postpartum. This result likely reflects the challenges that many women face in attending their postpartum visit in a timely manner. Not all women could be contacted to complete Follow-Up Survey #2 immediately after their scheduled 6-week postpartum visit. Those who completed the survey later could have biased our primary outcome of LARC use to be higher since it allowed participants more time to obtain LARC. However, any effect should have been distributed equally in both arms. Finally,
J.H. Tang et al. / Contraception xx (2014) xxx–xxx Table 1 Baseline characteristics of women in the study by computer-randomized assignment (intervention arm versus control arm) Characteristic
Age a 14–17 years 18–25 years 26–34 years 35–45 years Race/ethnicity a,b White/other not Hispanic Black not Hispanic Hispanic Country of birth b United States Mexico Other Insurancea,b Private Public Self-pay Prenatal clinic attended b Public Private Has trouble with necessities b Yes No Educationb Less than 12 years 12–15 years 16 or more years Religion b Protestant Catholic Other Relationship status b Married In relationship Sometimes/no relationship Parity a 1 2 3 or more Delivery mode a,b Vaginal Cesarean Gestational age at birth a Term Preterm (less than 37 weeks) Pregnancy intention at time of conceptionb Trying for pregnancy Not trying but not using contraception Not trying but using contraception Reported desire for another pregnancy within 2 years Yes No Number of children desired in the future 0 1 2 or more
Assigned to receive script (intervention arm: N=400), n (%)
Assigned to not receive script (control arm: N= 400), n (%)
13 (3.3) 157 (39.3) 179 (44.8) 51 (12.8)
21 (5.3) 174 (43.5) 166 (41.5) 39 (9.8)
102 (25.7) 139 (35.1) 156 (39.3)
92 (23.1) 145 (36.5) 160 (40.3)
227 (58.4) 109 (28.0) 53 (13.6)
223 (57.8) 104 (26.9) 59 (15.3)
103 (25.8) 156 (39.1) 140 (35.1)
87 (21.8) 174 (43.5) 139 (34.8)
294 (74.6) 100 (25.4)
311 (78.7) 84 (21.3)
104 (26.0) 296 (74.0)
108 (27.1) 290 (72.9)
128 (32.1) 196 (49.1) 75 (18.8)
140 (35.0) 194 (48.5) 66 (16.5)
225 (59.1) 115 (30.2) 41 (10.8)
235 (61.0) 121 (31.4) 29 (7.5)
156 (39.0) 176 (44.0) 68 (17.0)
142 (35.8) 186 (46.9) 69 (17.4)
153 (38.2) 119 (29.8) 128 (32.0)
165 (41.2) 121 (30.3) 114 (28.5)
313 (78.2) 87 (21.8)
321 (80.3) 78 (19.5)
370 (92.5) 30 (7.5)
370 (92.5) 30 (7.5)
129 (32.6) 190 (48.0) 77 (19.4)
148 (37.2) 175 (44.0) 75 (18.8)
74 (18.6) 324 (81.4)
78 (19.6) 321 (80.4)
191 (47.8) 90 (22.5) 119 (29.7)
153 (38.2) 139 (34.8) 108 (27.0)
(continued on next page)
5
Table 1 (continued) Characteristic
Received prenatal counseling about contraception b Yes No Prior use of long-acting reversible contraception Yes No Intent to use long-acting reversible contraception (pre-randomization) Yes No a b
Assigned to receive script (intervention arm: N= 400), n (%)
Assigned to not receive script (control arm: N=400), n (%)
274 (69.9) 118 (30.1)
293 (73.8) 104 (26.2)
46 (11.5) 354 (88.5)
41 (10.2) 359 (89.8)
157 (39.2) 243 (60.8)
146 (36.5) 254 (63.5)
Information obtained from medical record. Does not equal 100% because of non-response.
those participants who were randomized to the LARC script may have been more likely to have requested assistance from our research assistants to assist with scheduling their postpartum visit or viewed interest in LARC as a desirable answer, which could have increased LARC uptake and interest, respectively. Our results may not be generalizable to all U.S. populations. One third of the participants in our study had no health insurance, 40% were of Hispanic ethnicity and 40% were foreign born. A recent analysis showed that Latinas and foreign-born women are significantly more likely to use LARC than other populations in the U.S. [24]. We do not know if the LARC script would have been more or less effective in U.S. populations that do not already have relatively high LARC use. We also do not know what impact the script will have once the Affordable Care Act is implemented, and women with insurance no longer have to cost share for their contraception [25]. In conclusion, our study found that a brief postpartum educational intervention on LARC did not significantly increase LARC use in our population. However, it did significantly increase LARC interest, and it may increase LARC use in populations where other barriers to LARC are reduced. Although education is important, further research should identify these barriers and interventions to address them so that all women have access to LARC. Acknowledgment We wish to thank the study participants, WakeMed Health and Hospitals, Wake County Human Services and the following contributors from the University of North Carolina at Chapel Hill: Kathia Peña-Centeno (Fellowship in Family Planning #3631), Erika Campos (Fellowship in Family Planning #3631), Sandra Re, Bryna Waters, and Sarah Melvin. REDCap database management was supported by the North Carolina Translational and Clinical Sciences
6
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
Table 2 Primary and secondary outcomes for contraceptive use and interest by intervention arm (educational script versus no educational script) Outcome
Educational script N
Analysis by computer-randomized intervention (assigned intervention) Using long-acting reversible contraception 369 Interest in long-acting reversible contraception 369 Using long-acting reversible contraception Interested but not using long-acting reversible contraception Not interested in long-acting reversible contraception Using any contraceptive a 368 Vaginal delivery, using long-acting reversible contraception a,b 289 Cesarean delivery, using long-acting reversible contraception a,b 80
n
No educational script %
N
65
17.6
369 369
65 115 189 335 45 20
17.6 31.2 51.2 91.0 15.6 25.0
366 295 73
n
%
49
13.3
49 95 225 325 39 10
13.3 25.8 60.9 88.8 13.2 13.7
pValue
Relative risk (95% CI)
0.10 0.01
1.3 (0.9–1.9) N/A
0.32 0.42 0.08
1.0 (1.0–1.1) 1.2 (0.8–1.8) 1.8 (0.9–3.6)
N=total within each arm; n=total within each subgroup. a N does not equal 369 for assigned groups because of non-response. b p=.28, for test of interaction between “using long-acting reversible contraception” and “delivery mode”.
Institute at the University of North Carolina and Grant UL1RR025747 from the Clinical and Translational Science Award program of the Division of Research Resources, National Institutes of Health.
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.contraception.2014.03.017. References [1] Conde-Agudelo A, Rosas-Bermúdez A, Kafury-Goeta AC. Birth spacing and risk of adverse perinatal outcomes: a meta-analysis. JAMA 2006;295:1809–23. [2] U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2020. Available at: http://www.healthypeople.gov/2020/topicsobjectives2020/ objectiveslist.aspx?topicId=13 [Retrieved June 8, 2012]. [3] North Carolina State Center for Health Statistics. North Carolina Pregnancy Risk Assessment Monitoring System Survey results. Available at: http://www.schs.state.nc.us/SCHS/prams/2008/ BP_IUD.html 2008 [Retrieved September 29, 2010]. [4] Lopez LM, Hiller JE, Grimes DA. Education for contraceptive use by women after childbirth. Cochrane Database Syst Rev 2010;20(1): CD001863. [5] Trussell J. Choosing a contraceptive: efficacy, safety, and personal considerations. In: Hatcher RA, Trussell J, Nelson AL, Cates W, Stewart FH, & Kowal D, editors. Contraceptive Technology. New York, NY: Ardent Media; 2007, pp. 747–826. [6] ACOG Long-Acting Reversible Contraception Working Group. Increasing use of contraceptive implants and intrauterine devices to reduce unintended pregnancy. ACOG Committee Opinion No. 450. Obstet Gynecol 2009;114:1434–8. [7] Baldwin MK, Edelman AB. The effect of long-acting reversible contraception on rapid repeat pregnancy in adolescents: a review. J Adolesc Health 2013;52(4 Suppl):S47–53. [8] Thiel de Bocanegra H, Chang R, Menz M, Howell M, Darney P. Postpartum contraception in publicly funded programs and interpregnancy intervals. Obstet Gynecol 2013;122:296–303. [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 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] Winner B, Peipert JF, Zhao Q, Buckel C, Madden T, Allsworth JE, et al. Effectiveness of long-acting reversible contraception. NEJM 2012;366:1998–2007. [12] Secura GM, Allsworth JE, Madden T, Mullersman JL, Peipert JF. The Contraceptive CHOICE project: reducing barriers to long-acting reversible contraception. Am J Obstet Gynecol 2010;203:115.e1–7. [13] Schulz KF, Altman DG, Moher D, for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomized trials. BMJ 2010;340:698–702. [14] Fleiss JL, Levin B, Paik C. Statistical Methods for Rates and Proportions. 3rd ed. New York, NY: John Wiley & Sons, Inc.; 2003. [15] Koch GG, Edwards S. Clinical efficacy trials with categorical data. In: & Peace KE, editor. Biopharmaceutical Statistics for Drug Development. New York, NY: Marcel Dekker; 1988, pp. 403–51. [16] Zou GY. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6. [17] Miller ES, Grobman WA. Interbirth interval with frequency of cesarean delivery. Obstet Gynecol 2011;118:39–42. [18] Albrechtsen S, Rasmussen S, Dalaker K, Irgens LM. Reproductive career after breech presentation: subsequent pregnancy rates, interpregnancy interval, and recurrence. Obstet Gynecol 1998;92:345–50. [19] Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) — a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. [20] Bolam A, Manandhar DS, Shrestha P, Ellis M, de L Costello AM. The effects of postnatal health education for mothers on infant care and family planning practices in Nepal: a randomized controlled trial. BMJ 1998;316:805–11. [21] Saeed GA, Fakhar S, Rahim F, Tabassum S. Change in trend of contraceptive uptake — effect of education leaflets and counseling. Contraception 2008;77:377–81. [22] Gilliam M, Knight S, McCarthy Jr M. Success with oral contraceptives: a pilot study. Contraception 2004;69:413–8. [23] Proctor A, Jenkins TR, Loeb T, Elliot M, Ryan A. Patient satisfaction with 3 methods of postpartum contraceptive counseling: a randomized, prospective trial. J Reprod Med 2006;51:377–82. [24] 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. [25] Gee R. Preventive services for women under the Affordable Care Act. Obstet Gynecol 2012;120:12–4.
Original research article
Effect of an educational script on postpartum contraceptive use: a randomized controlled trial☆,☆☆,★ Jennifer H. Tang a,⁎, Rosalie C. Dominik b , Matthew L. Zerden a , Sarah B. Verbiest a , Seth C. Brody a, c , Gretchen S. Stuart a a
Department of Obstetrics and Gynecology, University of North Carolina, Chapel Hill, NC 27523, USA b Department of Biostatistics, University of North Carolina, Chapel Hill, NC 27523, USA c Department of Obstetrics and Gynecology, WakeMed Health and Hospitals, Raleigh, NC, USA Received 31 October 2013; revised 9 February 2014; accepted 23 March 2014
Abstract Objectives: Short interpregnancy intervals are associated with adverse perinatal outcomes. Increased postpartum use of long-acting reversible contraception (LARC) could reduce short interpregnancy intervals. Therefore, the primary objective of our study was to evaluate if a postpartum educational script about LARC (LARC script) could increase postpartum LARC utilization at the 6-week postpartum visit. Study design: We completed a randomized controlled trial (1:1) of 800 postpartum women in Raleigh, North Carolina. Women were recruited from the postpartum unit and randomized to receive (intervention) or not receive (control) the LARC script prior to hospital discharge. We conducted follow-up phone interviews to assess LARC use after their 6-week postpartum visit. We used Pearson's chi-squared test to compare LARC use between arms. Results: Between May 2011 and January 2012, 400 women were randomized to each arm. Three hundred sixty-nine women (92.3%) in each arm were successfully contacted after their 6-week postpartum visit. LARC use was reported by 17.6% and 13.3% of women in the intervention and control arms, respectively (p=.103). Conclusions: The LARC script did not increase LARC utilization at the 6-week postpartum visit. Future studies should evaluate interventions that address both educational and systematic barriers to postpartum LARC uptake. Implications: Use of a postpartum educational script focused on the intrauterine device and contraceptive implant did not increase their utilization after the 6-week postpartum visit. Future studies should evaluate interventions that address both educational and systematic barriers to postpartum contraceptive uptake. © 2014 Elsevier Inc. All rights reserved. Keywords: Contraception; Reversible; Patient education; Intrauterine device; Contraceptive implant
1. Introduction Short interpregnancy intervals are associated with adverse perinatal outcomes [1]. In the U.S. and North Carolina, 38% and 42% of pregnancies have interpregnancy intervals less
☆ Funding Acknowledgement: Funded by the Fellowship in Family Planning Grant #3631. ☆☆ Conflicts of Interest: The authors report no conflict of interest. ★ Clinical Trial Registration: ClinicalTrials.gov, www.clinicaltrials. gov, NCT01443533. ⁎ Corresponding author. 3031 Old Clinic Building, CB #7570, Chapel Hill, NC 27599-7570, USA. Tel.: +1-919-843-9546; fax: +1-919-843-9437. E-mail address: [email protected] (J.H. Tang).
http://dx.doi.org/10.1016/j.contraception.2014.03.017 0010-7824/© 2014 Elsevier Inc. All rights reserved.
than 18 months, respectively [2,3]. Postpartum contraceptive counseling may play a role in decreasing short interpregnancy intervals. A Cochrane review of eight randomized controlled trials found that postpartum contraceptive counseling can increase postpartum contraceptive use [4]. However, none of the trials focused on the most effective reversible contraceptives, long-acting reversible contraception (LARC) [5,6]. In the U.S., four LARC methods are available: a hormonal implant, a copper intrauterine device (IUD) and two levonorgestrel-releasing intrauterine systems. The American College of Obstetricians and Gynecologists has recommended increased LARC use to reduce the high rate of unintended pregnancy [6]. Increased postpartum use of LARC could also prevent short interpregnancy intervals
2
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
since they all provide effective contraception for at least 3 years [7,8]. Three main barriers prevent increased use of LARC: low patient awareness, high upfront costs and lack of health care provider knowledge or skills [5,9,10]. The Contraceptive CHOICE project overcame these barriers and achieved a LARC uptake rate of 77% in 7486 women by using three interventions: (1) a brief scripted introduction to LARC methods (LARC script), (2) removal of financial barriers in obtaining LARC and (3) ensuring that all eligible women interested in the IUD had a provider to place it [11,12]. These interventions could also increase LARC use in the postpartum population. However, the latter two interventions may not be easily replicable in other settings. The LARC script could be easily modified for the postpartum population. Therefore, the primary objective of this study was to determine if receiving a modified version of the LARC script in the immediate postpartum period would increase postpartum LARC use. We hypothesized that women who received our LARC script during their postpartum hospitalization would be more likely to report LARC use after their 6-week postpartum visit, compared to women who did not receive the script. Secondary objectives were to assess if the script would increase overall contraceptive use or interest in using a LARC method after the 6-week postpartum visit.
2. Materials and methods This study was a parallel (1:1) randomized controlled trial of 800 women admitted to the postpartum unit of WakeMed Hospital between May 2011 and January 2012. The study protocol followed CONSORT guidelines and was approved by the Institutional Review Boards of the University of North Carolina and WakeMed Hospital [13]. WakeMed Hospital is a large private nonprofit hospital in Raleigh, North Carolina, with about 5000 deliveries annually. Immediate postpartum insertion of LARC was not available there during the study period, and no standard procedures for postpartum contraceptive counseling existed. The research assistants looked through the delivery log every weekday morning and recruited women who were likely to be discharged first. Due to involvement in other studies, they could not recruit consecutively every day and instead recruited until they reached a minimum quota of 25 randomized women per week. The eligibility criteria for the study were as follows: (1) admission to the postpartum unit, (2) delivery of a live infant N 24 weeks gestational age, (3) age 14–45 years, (4) ability to speak English or Spanish fluently, (5) willing to be contacted by phone until at least 8 weeks after delivery, (6) no history of a tubal ligation or hysterectomy, (7) current partner has not had a vasectomy, (8) no use of fertility treatment to conceive this pregnancy and (9) no previous randomization into the study.
The research assistants approached eligible women in their postpartum hospital rooms. Women were told that the study's purpose was to learn about contraceptive use among postpartum women and they would be randomized to receive or not receive an experimental educational message. Women underwent informed consent if they agreed to participate and completed a 20-question Baseline Survey via face-to-face interviews. Additional demographic and obstetric history data were abstracted from their medical records. After each interview, the research assistant opened a sequentially numbered opaque sealed envelope. The envelope contained a piece of paper that was either blank (control arm) or had the 1-min LARC script (Appendix A) in English and Spanish (intervention arm). Participants allocated to the intervention arm had the LARC script read aloud to them and were handed the script. If the participant wanted more information about a LARC method, she was given company promotional information approved by the Food and Drug Administration for that method. The research assistants helped the participants schedule their 6-week postpartum visits (the standard of care at the hospital) as needed. Data were then collected from two follow-up phone surveys. Follow-Up Survey #1 occurred 1–2 weeks postpartum and consisted of a 1-min questionnaire to confirm the participant's 6-week postpartum visit date. Phone calls to complete Follow-Up Survey #2 began the day after the participants' scheduled 6-week postpartum visit and involved a 10-min survey with 25 questions. Follow-Up Survey #2 included questions about whether or not the participant was using LARC (primary outcome), whether or not she was interested in using LARC (secondary outcome) and which contraceptive method she was using (secondary outcome). To assess if a participant was interested in using LARC, we asked, “Are you interested in using the intrauterine device (the IUD) or the birth control implant (Implanon) for birth control?” If the participant answered “Yes”, she was then asked “Which one are you interested in using?” At the end of Follow-Up Survey #2, all participants were given a chance to learn more about LARC methods by having the LARC script read to them. A multi-tiered contact system was used to track the subjects. If the participant could not be reached to complete Follow-Up Survey #2 after five attempts to her primary or secondary phone numbers, she was emailed and/or paged. If there still was no response, the participant's contact persons, postpartum clinic and pediatric clinic were contacted if she had given permission to contact them during enrollment. Finally, if all previous attempts failed and the participant was still within 3 months of her scheduled 6-week postpartum visit, the survey was mailed to the patient, with a prestamped and pre-addressed return envelope. The participant was considered lost-to-follow-up if Follow-Up Survey #2 was not returned within 1 month of being sent. Simple 1:1 randomization was achieved via computergenerated randomly permuted block randomization, with block sizes of 6, 8 and 10. The randomization scheme was
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
developed and stored by an independent statistician at the North Carolina Translational and Clinical Sciences Institute and used to prepare the allocation envelopes. The study investigators were blinded to the randomization until data analysis. The potential for bias was decreased by designating that the research assistant who opened the allocation envelope for a participant was not the same research assistant who completed that participant's Follow-Up Surveys. For our sample size calculations, we assumed that 20% of women in the control arm would be using LARC after their 6-week postpartum visit, based on LARC use rates reported among women 2–3 months postpartum in the North Carolina Pregnancy Risk Assessment Monitoring Survey [3]. An analysis population with at least 626 women (313 in each arm) was required to provide 80% power (using a two-sided alpha of 0.05) to detect an absolute 10% increase to 30% in LARC use in the intervention arm [14]. Anticipating a maximum drop-out rate of 20% at the time of Follow-Up Survey #2, we planned to randomize 800 participants. Descriptive statistics were used to summarize the baseline variables by assigned arm. For analysis of LARC use and any contraceptive use after the 6-week postpartum visit, we used Pearson's chi-square test to compare the proportions of use between the two arms and calculated the risk ratio (RR) and their 95% confidence intervals (CIs). Interest in a LARC method was considered an ordinal variable with three categories: using a LARC method, interested but not using a LARC method and not interested and not using a LARC method. The row mean score chi-square test was used to compare the distributions of the ordinal LARC interest variable between study arms [15]. During data analysis, we found that 34 participants were not allocated to their randomly assigned arm due to error in preparing the allocation envelopes. Therefore, we first performed all planned analyses using the arm assigned by the computergenerated randomization. We then performed sensitivity analyses using the actual arm allocated to the participants. We repeated the LARC use comparison by mode of delivery and used modified Poisson regression to assess interaction between mode of delivery and intervention [16]. Based on prior studies, our hypothesis was that women who had cesarean deliveries would desire longer birth intervals than women who delivered vaginally [17,18]. Differences between the two randomized arms were tested at the 0.05 alpha level, whereas testing for the interaction between mode of delivery and the randomized arm was performed at the 0.10 alpha level. All analyses were done using Stata 12.0 (StataCorp LP, College Station, TX). Study data were managed using the Web-based REDCap (Research Electronic Data Capture) electronic data system [19].
3. Results The study was conducted from May 2011 through April 2012. Nine hundred eighty-nine women were screened, 800
3
(80.9%) enrolled and 738 (92.3%) of the enrolled women completed the study (Fig. 1). The computer-generated randomization sequence assigned 400 women to receive the LARC script (intervention arm) and 400 women to receive no LARC script (control arm). Due to an error in preparing the allocation envelopes, 394 women were actually allocated to the LARC script (15 incorrectly), and 403 women were allocated to no LARC script (19 incorrectly). Participants in both arms had similar baseline characteristics, except that women in the intervention arm were more likely to desire no future children (47.8%) than in the control arm (38.2%) (Table 1). Only a third were trying for pregnancy at the time of conception, and about 80% did not want another pregnancy within 2 years. Approximately 10% had previously used LARC, and almost 40% intended to use LARC as postpartum contraception. Of the 738 women who completed Follow-Up Survey #2, 620 (84.0%) reported attending a postpartum visit. The median time between the 6-week postpartum visit and Follow-up Survey #2 was 11 days, with a range 0–121 days. In the intervention arm, 17.6% reported using a LARC method after their 6-week postpartum visit, compared to 13.3% in the control arm (RR 1.3; 95% CI 0.9–1.9; p=.10) (Table 2). Results of the sensitivity analysis using actual allocation were similar, with no significant difference between the two arms. We found a significant difference in LARC interest between the two arms (Table 2). Thirty one percent of women in the intervention arm and 26% in the control arm were interested in but not yet using LARC. The comparison of the ordinal interest variable between the two arms showed significantly increased interest in the intervention arm compared to the control arm (p=.01). Sensitivity analysis using actual allocation revealed similar results. Of note, in the intervention arm, 49% of participants were either using or interested in using LARC, compared to 39% in the control arm. No statistically significant difference was found between arms for use of any contraceptive method (RR 1.0; 95% CI 1.0–1.1; p=.32) (Table 2). The interaction between mode of delivery and script was also not statistically significant (p=.28).
4. Discussion In our study, administration of a LARC script to postpartum women did not significantly increase postpartum LARC use. However, the script significantly increased interest in using LARC. The difference between the effect of the script on LARC interest and LARC use may suggest that many participants interested in using LARC were unable to obtain it. Prior to randomization in the study, almost 40% of participants planned to use LARC for postpartum contraception, but less than half were actually using it after their 6-week postpartum visit. Barriers to receiving LARC, such
4
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
Fig. 1. Study flow chart.
as delayed timing of the postpartum visit, likely exist and should be further investigated. Four previous randomized controlled trials have evaluated the impact of immediate postpartum contraceptive education on postpartum contraceptive use. Two studies found that the educational intervention significantly increased overall postpartum contraceptive use, one of which also led to a 10% increase in postpartum IUD use [20,21]. The educational interventions in these studies were each 20-min long, but neither was focused on LARC nor was they performed in the U.S. The other two studies were performed in the U.S. but were inconclusive [22,23]. The first study was underpowered to show a difference in contraceptive use, and the second only assessed satisfaction with the educational intervention. Differences in study populations and types of interventions may account for the contrasting results with our study, which found that neither postpartum IUD nor overall contraceptive use increased with our educational intervention. If future studies in the U.S. are
performed, they should consider using longer educational counseling sessions. Our study's strengths include its large size, blinded follow-up and randomization of the study intervention. Although we discovered an error in our randomization assignment, sensitivity analyses confirmed minimum impact on the primary and secondary outcomes. A potential weakness is that we did not complete Follow-Up Survey #2 at the same postpartum interval for all participants. Some participants had their 6-week postpartum visits scheduled as late as 10 weeks postpartum. This result likely reflects the challenges that many women face in attending their postpartum visit in a timely manner. Not all women could be contacted to complete Follow-Up Survey #2 immediately after their scheduled 6-week postpartum visit. Those who completed the survey later could have biased our primary outcome of LARC use to be higher since it allowed participants more time to obtain LARC. However, any effect should have been distributed equally in both arms. Finally,
J.H. Tang et al. / Contraception xx (2014) xxx–xxx Table 1 Baseline characteristics of women in the study by computer-randomized assignment (intervention arm versus control arm) Characteristic
Age a 14–17 years 18–25 years 26–34 years 35–45 years Race/ethnicity a,b White/other not Hispanic Black not Hispanic Hispanic Country of birth b United States Mexico Other Insurancea,b Private Public Self-pay Prenatal clinic attended b Public Private Has trouble with necessities b Yes No Educationb Less than 12 years 12–15 years 16 or more years Religion b Protestant Catholic Other Relationship status b Married In relationship Sometimes/no relationship Parity a 1 2 3 or more Delivery mode a,b Vaginal Cesarean Gestational age at birth a Term Preterm (less than 37 weeks) Pregnancy intention at time of conceptionb Trying for pregnancy Not trying but not using contraception Not trying but using contraception Reported desire for another pregnancy within 2 years Yes No Number of children desired in the future 0 1 2 or more
Assigned to receive script (intervention arm: N=400), n (%)
Assigned to not receive script (control arm: N= 400), n (%)
13 (3.3) 157 (39.3) 179 (44.8) 51 (12.8)
21 (5.3) 174 (43.5) 166 (41.5) 39 (9.8)
102 (25.7) 139 (35.1) 156 (39.3)
92 (23.1) 145 (36.5) 160 (40.3)
227 (58.4) 109 (28.0) 53 (13.6)
223 (57.8) 104 (26.9) 59 (15.3)
103 (25.8) 156 (39.1) 140 (35.1)
87 (21.8) 174 (43.5) 139 (34.8)
294 (74.6) 100 (25.4)
311 (78.7) 84 (21.3)
104 (26.0) 296 (74.0)
108 (27.1) 290 (72.9)
128 (32.1) 196 (49.1) 75 (18.8)
140 (35.0) 194 (48.5) 66 (16.5)
225 (59.1) 115 (30.2) 41 (10.8)
235 (61.0) 121 (31.4) 29 (7.5)
156 (39.0) 176 (44.0) 68 (17.0)
142 (35.8) 186 (46.9) 69 (17.4)
153 (38.2) 119 (29.8) 128 (32.0)
165 (41.2) 121 (30.3) 114 (28.5)
313 (78.2) 87 (21.8)
321 (80.3) 78 (19.5)
370 (92.5) 30 (7.5)
370 (92.5) 30 (7.5)
129 (32.6) 190 (48.0) 77 (19.4)
148 (37.2) 175 (44.0) 75 (18.8)
74 (18.6) 324 (81.4)
78 (19.6) 321 (80.4)
191 (47.8) 90 (22.5) 119 (29.7)
153 (38.2) 139 (34.8) 108 (27.0)
(continued on next page)
5
Table 1 (continued) Characteristic
Received prenatal counseling about contraception b Yes No Prior use of long-acting reversible contraception Yes No Intent to use long-acting reversible contraception (pre-randomization) Yes No a b
Assigned to receive script (intervention arm: N= 400), n (%)
Assigned to not receive script (control arm: N=400), n (%)
274 (69.9) 118 (30.1)
293 (73.8) 104 (26.2)
46 (11.5) 354 (88.5)
41 (10.2) 359 (89.8)
157 (39.2) 243 (60.8)
146 (36.5) 254 (63.5)
Information obtained from medical record. Does not equal 100% because of non-response.
those participants who were randomized to the LARC script may have been more likely to have requested assistance from our research assistants to assist with scheduling their postpartum visit or viewed interest in LARC as a desirable answer, which could have increased LARC uptake and interest, respectively. Our results may not be generalizable to all U.S. populations. One third of the participants in our study had no health insurance, 40% were of Hispanic ethnicity and 40% were foreign born. A recent analysis showed that Latinas and foreign-born women are significantly more likely to use LARC than other populations in the U.S. [24]. We do not know if the LARC script would have been more or less effective in U.S. populations that do not already have relatively high LARC use. We also do not know what impact the script will have once the Affordable Care Act is implemented, and women with insurance no longer have to cost share for their contraception [25]. In conclusion, our study found that a brief postpartum educational intervention on LARC did not significantly increase LARC use in our population. However, it did significantly increase LARC interest, and it may increase LARC use in populations where other barriers to LARC are reduced. Although education is important, further research should identify these barriers and interventions to address them so that all women have access to LARC. Acknowledgment We wish to thank the study participants, WakeMed Health and Hospitals, Wake County Human Services and the following contributors from the University of North Carolina at Chapel Hill: Kathia Peña-Centeno (Fellowship in Family Planning #3631), Erika Campos (Fellowship in Family Planning #3631), Sandra Re, Bryna Waters, and Sarah Melvin. REDCap database management was supported by the North Carolina Translational and Clinical Sciences
6
J.H. Tang et al. / Contraception xx (2014) xxx–xxx
Table 2 Primary and secondary outcomes for contraceptive use and interest by intervention arm (educational script versus no educational script) Outcome
Educational script N
Analysis by computer-randomized intervention (assigned intervention) Using long-acting reversible contraception 369 Interest in long-acting reversible contraception 369 Using long-acting reversible contraception Interested but not using long-acting reversible contraception Not interested in long-acting reversible contraception Using any contraceptive a 368 Vaginal delivery, using long-acting reversible contraception a,b 289 Cesarean delivery, using long-acting reversible contraception a,b 80
n
No educational script %
N
65
17.6
369 369
65 115 189 335 45 20
17.6 31.2 51.2 91.0 15.6 25.0
366 295 73
n
%
49
13.3
49 95 225 325 39 10
13.3 25.8 60.9 88.8 13.2 13.7
pValue
Relative risk (95% CI)
0.10 0.01
1.3 (0.9–1.9) N/A
0.32 0.42 0.08
1.0 (1.0–1.1) 1.2 (0.8–1.8) 1.8 (0.9–3.6)
N=total within each arm; n=total within each subgroup. a N does not equal 369 for assigned groups because of non-response. b p=.28, for test of interaction between “using long-acting reversible contraception” and “delivery mode”.
Institute at the University of North Carolina and Grant UL1RR025747 from the Clinical and Translational Science Award program of the Division of Research Resources, National Institutes of Health.
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.contraception.2014.03.017. References [1] Conde-Agudelo A, Rosas-Bermúdez A, Kafury-Goeta AC. Birth spacing and risk of adverse perinatal outcomes: a meta-analysis. JAMA 2006;295:1809–23. [2] U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2020. Available at: http://www.healthypeople.gov/2020/topicsobjectives2020/ objectiveslist.aspx?topicId=13 [Retrieved June 8, 2012]. [3] North Carolina State Center for Health Statistics. North Carolina Pregnancy Risk Assessment Monitoring System Survey results. Available at: http://www.schs.state.nc.us/SCHS/prams/2008/ BP_IUD.html 2008 [Retrieved September 29, 2010]. [4] Lopez LM, Hiller JE, Grimes DA. Education for contraceptive use by women after childbirth. Cochrane Database Syst Rev 2010;20(1): CD001863. [5] Trussell J. Choosing a contraceptive: efficacy, safety, and personal considerations. In: Hatcher RA, Trussell J, Nelson AL, Cates W, Stewart FH, & Kowal D, editors. Contraceptive Technology. New York, NY: Ardent Media; 2007, pp. 747–826. [6] ACOG Long-Acting Reversible Contraception Working Group. Increasing use of contraceptive implants and intrauterine devices to reduce unintended pregnancy. ACOG Committee Opinion No. 450. Obstet Gynecol 2009;114:1434–8. [7] Baldwin MK, Edelman AB. The effect of long-acting reversible contraception on rapid repeat pregnancy in adolescents: a review. J Adolesc Health 2013;52(4 Suppl):S47–53. [8] Thiel de Bocanegra H, Chang R, Menz M, Howell M, Darney P. Postpartum contraception in publicly funded programs and interpregnancy intervals. Obstet Gynecol 2013;122:296–303. [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 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] Winner B, Peipert JF, Zhao Q, Buckel C, Madden T, Allsworth JE, et al. Effectiveness of long-acting reversible contraception. NEJM 2012;366:1998–2007. [12] Secura GM, Allsworth JE, Madden T, Mullersman JL, Peipert JF. The Contraceptive CHOICE project: reducing barriers to long-acting reversible contraception. Am J Obstet Gynecol 2010;203:115.e1–7. [13] Schulz KF, Altman DG, Moher D, for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomized trials. BMJ 2010;340:698–702. [14] Fleiss JL, Levin B, Paik C. Statistical Methods for Rates and Proportions. 3rd ed. New York, NY: John Wiley & Sons, Inc.; 2003. [15] Koch GG, Edwards S. Clinical efficacy trials with categorical data. In: & Peace KE, editor. Biopharmaceutical Statistics for Drug Development. New York, NY: Marcel Dekker; 1988, pp. 403–51. [16] Zou GY. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6. [17] Miller ES, Grobman WA. Interbirth interval with frequency of cesarean delivery. Obstet Gynecol 2011;118:39–42. [18] Albrechtsen S, Rasmussen S, Dalaker K, Irgens LM. Reproductive career after breech presentation: subsequent pregnancy rates, interpregnancy interval, and recurrence. Obstet Gynecol 1998;92:345–50. [19] Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) — a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. [20] Bolam A, Manandhar DS, Shrestha P, Ellis M, de L Costello AM. The effects of postnatal health education for mothers on infant care and family planning practices in Nepal: a randomized controlled trial. BMJ 1998;316:805–11. [21] Saeed GA, Fakhar S, Rahim F, Tabassum S. Change in trend of contraceptive uptake — effect of education leaflets and counseling. Contraception 2008;77:377–81. [22] Gilliam M, Knight S, McCarthy Jr M. Success with oral contraceptives: a pilot study. Contraception 2004;69:413–8. [23] Proctor A, Jenkins TR, Loeb T, Elliot M, Ryan A. Patient satisfaction with 3 methods of postpartum contraceptive counseling: a randomized, prospective trial. J Reprod Med 2006;51:377–82. [24] 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. [25] Gee R. Preventive services for women under the Affordable Care Act. Obstet Gynecol 2012;120:12–4.
