Abstract
Study Objective:
To determine the association of postpartum contraceptive use with repeat deliveries among adolescents and youth.
Design:
Retrospective, observational analysis of electronic health record data.
Setting:
Single, urban facility in Denver, Colorado, United States.
Participants:
Women aged 10-24 years who gave birth between January 1, 2011 and December 31, 2015.
Interventions and Main Outcome Measures:
Postpartum contraceptive use and time to subsequent delivery.
Results:
Among 4068 women, 1735 (43%) used postpartum contraception. In adjusted analyses, characteristics associated with contraceptive use included Hispanic ethnicity (relative risk [RR], 1.1; P = .03), incremental prenatal visits (RR, 1.01; P = .047), and attendance at postpartum care (RR, 1.60; P < .001). Long-acting reversible contraceptive (LARC) use was higher among women younger than 15 years (reference: 20-24 years; RR, 1.12; P < .001) and lower among women aged 18-19 years (RR, 0.93; P = .009). Hispanic women had higher rates of LARC use than non-Hispanic women (RR, 1.07; P = .02). Compared with inpatient LARC placement, outpatient placement (1-4 weeks and 5 or more weeks) rates were lower (RR, 0.77 and RR, 0.89, respectively; P < .001). Time to subsequent delivery was shorter in non-LARC users (median, 659 days) and contraception nonusers (median, 624 days) compared with LARC users (median, 790 days; P < .001); non-LARC postpartum contraceptive use did not significantly alter time to repeat delivery compared with that in women who used no method (P = .24).
Conclusion:
Postpartum LARC use reduced the risk of repeat pregnancy with a significant increase in time to the next delivery. Non-LARC use was not different from no contraceptive use in terms of time to repeat delivery.
Keywords: Adolescent pregnancy, Postpartum contraception, Interpregnancy interval, Repeat delivery
Introduction
Adolescent pregnancy is associated with poor pregnancy outcomes and a high risk of shortened interval to repeat pregnancy.1–3 To avoid adverse pregnancy outcomes, the World Health Organization recommends 24 months for proper pregnancy spacing.4 In the United States, at least 18 months between a birth and subsequent conception is recommended.5
Previous research on postpartum contraception and its association with subsequent pregnancy has been conducted in Colorado.6 A targeted public health intervention, the Colorado Family Planning Initiative (CFPI), was implemented in Colorado in 2009.7,8 The objective of this initiative was to finance provider training in the counseling, placement, operational support, and provision of long-acting reversible contraceptives (LARCs; ie, intrauterine device or implant) in an effort to reduce the number of adolescent pregnancies.7,8 The program was associated with a reduced rate of births among adolescents overall, as well as second and higher-order pregnancies in the population.7,8 Additionally, using the Colorado birth certificate registry, a previous analysis showed that among an adolescent population, a small proportion of women (44%) achieved the recommended 18-month interpregnancy interval.9 Before the CFPI was implemented, 42% of women had an optimal interval compared with 47% after the program was initiated, which resulted in an increased odds of an appropriate interpregnancy interval (odds ratio [confidence interval] = 1.2 [1.1-1.2]).9
On the basis of these findings, we hypothesized that a gap in use of postpartum contraception would be associated with shortened pregnancy spacing in the Denver adolescent and young adult populations. We also hypothesized that young women using highly effective methods of postpartum contraception would be less likely to have a repeat delivery. Our aims for this analysis were to identify prevalence of postpartum contraceptive use overall and according to subtype (LARC vs non-LARC) and to observe the association of postpartum contraceptive use (and type) with time to repeat delivery. Non-LARC refers to any modern contraceptive method other than intrauterine devices and implants.10,11 As such, our primary outcome for this analysis was initiation of postpartum contraceptive use (stratified according to LARC vs non-LARC) and our secondary outcome was time to subsequent delivery.
Materials and Methods
This retrospective, observational analysis of electronic health record data was part of a regional distributed data network (ie, Colorado Health Observation Regional Data Service [CHORDS]), which supports public health evaluation and monitoring efforts. CHORDS standardizes aggregate medical and behavioral health data from contributing Colorado health care partners into a deidentified data set.12 Only 1 center’s data (Denver Health) was used for this analysis. Denver Health is the safety net provider for more than 25% of Denver’s total population, and cares for approximately 45% of all women 10-24 years of age in the metro area.
Sociodemographic data (ie, maternal age [at time of index delivery], race, and ethnicity) for participants were available from CHORDS. Antepartum (ie, onset of care and number of visits), intrapartum (ie, type of delivery [vaginal or cesarean] and outcome [live or stillbirth], postpartum visit (yes/no), and contraceptive information (ie, evidence of, type [LARC, non-LARC, or none], and onset date) was identified using CHORDS data and International Classification of Diseases (ICD), ninth revision, and ICD 10th revision codes (Appendix A). Women were included in the analysis if they were aged between 10 and 24 years at the delivery of their index pregnancy, which occurred (live or stillbirth of a singleton; twins were excluded) between January 1, 2011 and December 31, 2015.
Our primary outcome was initiation of any postpartum contraception within 3 months of delivery of the index birth. First, we compared users of any postpartum contraception with nonusers with descriptive statistics and bivariate analyses using χ2 and Wilcoxon rank sum testing. Then, among users of postpartum contraception, we compared those who chose LARC vs non-LARC methods. Risk factors significant to P less than .05 were included in multivariable analyses using a Poisson regression with robust error variance because postpartum contraception use was a relatively common outcome in our cohort.13
Our secondary outcome was time to repeat delivery within 33 months (1010 days since the index birth), which represents a pregnancy within 24 months with 280 days added for a 40-week gestation. Women were considered to have a repeat delivery if there was evidence of a subsequent birth in the hospital records within the observation interval (1010 days; 33 months) after their index birth. If there was no evidence of a delivery in the medical record, we assumed that these women had not yet had a repeat delivery and their data were censored at 1010 days. We calculated the interval from index delivery to subsequent delivery because date of conception was unavailable and was not computable on the basis of data available in the CHORDS. For our secondary analysis, we used a Cox proportional hazards model to examine the association of contraceptive method with time to subsequent delivery adjusted for covariates (ie, age, ethnicity, postnatal care). Analyses were performed using SAS version 9.4M3 (SAS Institute Inc, Cary NC). This analysis was approved by the Colorado Multiple Institutional Review Board (COMIRB 17-0940).
Results
Among 4068 adolescent and young adult women with index births in the study time frame, 1735 (1735/4068, 43%) women used postpartum contraception. Among those, 1454 (1454/1735, 84%) used LARC and 281 (281/1735,16%) used non-LARC methods. The remaining 2333 women (2333/4068, 57%) had no evidence of having received postpartum contraception (Fig. 1). Of all women in the cohort, (3404/4068, 84% did not have a repeat delivery in the observation interval.
Fig. 1.
Diagram of contraception and repeat pregnancy outcomes among a cohort of Denver Health women 10-24 years old, after index delivery, Denver, Colorado, 2011-2015. LARC, long-acting reversible contraceptive.
Most of the women in the study cohort were aged 20-24 years (2708/4068, 66.6%), identified as white (3452/4068, 85%) and Hispanic (3050/4068, 75%), had a median of 10 antenatal care visits (range, 0-28 visits), gave birth vaginally (3476/4068, 86%), experienced a live birth (4038/4068, 99%), and presented for postpartum care (2500/4068, 62%; Table 1). Among the 16% (664/4068) of women with a second documented delivery, 8% (56/664) were within 12 months, 25% (164/664) between 12 and 18 months, 28% (186/664) between 18 and 24 months, and 39% (258/664) between 25 and 33 months.
Table 1.
Comparison of Previously Delivered Adolescent and Young Adult Women According to Postpartum Contraception Use within 3 Months of Index Birth, Denver, Colorado, 2011-2015
| Characteristic | Postpartum Birth Control Method Initiated, n (%) |
P | ||
|---|---|---|---|---|
| Total |
No |
Yes |
||
| (N = 4068) | (n = 2333; 57.4%) | (n = 1735; 42.6%) | ||
| Age, years | .35 | |||
| <15 | 20 (0.5) | 10 (0.4) | 10 (0.6) | |
| 15-17 | 490 (12.1) | 273 (11.7) | 217 (12.5) | |
| 18-19 | 850 (20.9) | 471 (20.2) | 379 (21.8) | |
| 20-24 | 2708 (66.6) | 1579 (67.7) | 1129 (65.1) | |
| Race | .07 | |||
| White | 3452 (84.9) | 1955 (83.8) | 1497 (86.3) | |
| Black | 445 (10.9) | 269 (11.5) | 176 (10.1) | |
| Asian/other | 171 (4.2) | 109 (4.7) | 62 (3.6) | |
| Ethnicity | .008 | |||
| Hispanic | 3050 (75.0) | 1708 (73.2) | 1342 (19.2) | |
| Non-Hispanic | 851 (20.9) | 518 (22.2) | 333 (77.4) | |
| Unknown | 167 (4.1) | 107 (4.6) | 60 (3.5) | |
| Median number of antenatal care visits (range) | 10 (0-28) | 10 (0-26) | 11 (0-28) | <.001 |
| Mode of delivery | .95 | |||
| Vaginal | 3476 (85.8) | 1996 (85.8) | 1480 (85.8) | |
| Cesarean | 576 (14.2) | 330 (14.2) | 246 (14.3) | |
| Index pregnancy neonatal outcome | .08 | |||
| Live birth | 4038 (99.3) | 2311 (99.1) | 1727 (99.5) | |
| Stillbirth | 30 (0.7) | 22 (0.9) | 8 (0.5) | |
| Postpartum visit | 2500 (61.5) | 1245 (53.4) | 1255 (72.3) | <.001 |
| Subsequent delivery | 664 (16.3) | 488 (20.9) | 176 (10.1) | <.001 |
| n = 664 | n = 488 | n = 176 | ||
| Time to subsequent delivery | <.001 | |||
| <12 months | 56 (8.4) | 47 (9.6) | 9 (5.1) | |
| 12-18 months | 164 (24.7) | 142 (29.1) | 22 (12.5) | |
| 19-24 months | 186 (28.0) | 137 (28.1) | 49 (27.8) | |
| 25-33 months | 258 (38.9) | 162 (33.2) | 96 (54.6) | |
Age, race, ethnicity, antenatal care visits, neonatal outcome, and postpartum visit variables had 0% missing data; 0.4% of data were missing for mode of delivery.
Bivariate comparisons of women with documentation of postpartum contraceptive use within 3 months compared with those with no such documentation are presented in Table 1. Covariates significantly associated with contraceptive initiation included ethnicity (44% of Hispanic women compared with 39% of non-Hispanic women, 36% of those with unknown ethnicity; P = .008). Women who attended more prenatal visits were more likely to initiate any postpartum contraceptive method; among those who used contraception they had a median number of visits of 11 (range, 0-28 visits), and among those who did not use a method, they had a median of 10 visits (range, 0-26 visits; P < .001). Women who started a contraceptive method were more likely to have attended a postpartum visit (50% vs 31%; P < .001).
In the multivariable model (Table 2), we found that women who initiated postpartum contraception were more likely to be Hispanic (relative risk [RR], 1.11; P = .03), to attend a postpartum visit (RR, 1.60; P < .001), and with each increasing prenatal visit were more likely to initiate postpartum contraception (RR, 1.01; P = .047).
Table 2.
Multivariable Model: Relative Risk of Initiating Any Postpartum Contraception Method among Previously Delivered Adolescent and Young Adult Women, Denver, Colorado, 2011-2015
| Variable | Relative Risk | P |
|---|---|---|
| Hispanic ethnicity (reference: non-Hispanic) | 1.11 | .03 |
| Antenatal care visits for index pregnancy, 1 visit change | 1.01 | .047 |
| Attended postpartum visit (reference: no) | 1.60 | <.001 |
Women with unknown ethnicity were excluded from the multivariable analysis.
Table 3 shows women who had ICD codes suggestive of initiation of a postpartum contraceptive method within 3 months of delivery of their index birth stratified according to LARC or another (non-LARC) contraceptive method. In bivariate comparisons of these populations, LARC users were more likely to be age 17 years or younger at their index birth (85% vs 15%; P = .004), Hispanic (85% vs 15%; P = .04), and received a birth control method in the inpatient setting (92% LARC [inpatient] vs 70% LARC [1-4 weeks, outpatient setting] vs 81% LARC [5 or more weeks, outpatient setting]; P < .001).
Table 3.
Comparison of Previously Delivered Adolescent and Young Adult Women Who Initiated Postpartum Contraception within 3 Months of an Index Birth According to LARC vs Non-LARC Method, Denver, Colorado, 2011-2015
| Characteristic | Non-LARC |
LARC |
P |
|---|---|---|---|
| n = 281 | n = 1454 | ||
| Age, years | .004 | ||
| <15 | 0 (0.0) | 10 (0.7) | |
| 15-17 | 33 (11.7) | 184 (12.7) | |
| 18-19 | 83 (29.5) | 296 (20.4) | |
| 20-24 | 165 (58.7) | 964 (66.3) | |
| Race | .17 | ||
| White | 233 (82.9) | 1264 (86.9) | |
| Black | 37 (13.2) | 139 (9.6) | |
| Asian/other | 11 (3.9) | 51 (3.5) | |
| Ethnicity | .04 | ||
| Hispanic | 201 (71.5) | 1141 (78.5) | |
| Non-Hispanic | 69 (24.6) | 264 (18.2) | |
| Unknown | 11 (3.9) | 49 (3.4) | |
| Median number of antenatal care visits (range) | 11 (0-24) | 11 (0-28) | .19 |
| Mode of delivery | .10 | ||
| Vaginal | 249 (88.9) | 1231 (85.1) | |
| Cesarean | 31 (11.1) | 215 (14.9) | |
| Index pregnancy neonatal outcome | .60 | ||
| Live birth | 279 (99.3) | 1448 (99.6) | |
| Stillbirth | 2 (0.7) | 6 (0.4) | |
| Postpartum visit | 211 (75.1) | 1044 (71.8) | .26 |
| Time to receipt of birth control method | <.001 | ||
| Inpatient | 47 (16.7) | 528 (36.3) | |
| 1-4 weeks | 47 (16.7) | 107 (7.4) | |
| 5 or more weeks | 187 (66.6) | 819 (56.3) | |
| Subsequent delivery | 66 (23.5) | 110 (7.6) | <.001 |
| n = 66 | n = 110 | ||
| Time to subsequent delivery | <.001 | ||
| <12 months | 4 (6.1) | 5 (4.6) | |
| 12-18 months | 15 (22.7) | 7 (6.4) | |
| 19-24 months | 24 (36.4) | 25 (22.7) | |
| 25-33 months | 23 (34.9) | 73 (66.4) | |
Data are presented as n (%) except where otherwise noted. For the variable, “Mode of Delivery,” 0.4% of data were missing; there were no missing data in other variables.
LARC, long-acting reversible contraception.
In the multivariable model (Table 4), compared with young women aged 20-24 years, very young adolescents younger than the age of 15 years were more likely to initiate LARC (RR, 1.12; P < .001), whereas older adolescents aged 18-19 years were less likely to initiate LARC (RR, 0.93; P = .009). Compared with non-Hispanic women, Hispanic women were more likely to initiate LARC (RR, 1.07; P = .02); an interaction term between age and ethnicity was not significant. Women were more likely to initiate LARC in the inpatient setting compared with postpartum (an interaction term between age and time to initiation of birth control was not significant); women were less likely to initiate LARC within the 1- to 4-week time frame (RR, 0.77; P < .001), and after 5 weeks (RR, 0.89; P < .001).
Table 4.
Multivariable Model: Risk of Initiating Long-acting Reversible Contraceptive
| Variable | Relative Risk | P |
|---|---|---|
| Age (reference: 20 - 24), years | ||
| <15 | 1.12 | <.001 |
| 15-17 | 0.98 | .51 |
| 18-19 | 0.93 | .009 |
| Hispanic ethnicity (reference: non-Hispanic) | 1.07 | .02 |
| Time to receipt of birth control method (reference: inpatient) | ||
| 1-4 weeks | 0.77 | <.001 |
| 5 or more weeks | 0.89 | <.001 |
Women with unknown ethnicity were excluded from the multivariable analysis.
Time from index birth to subsequent delivery is shown in Figure 2. LARC use delayed the time to subsequent delivery compared with non-LARC use (hazard ratio [HR] 0.29; P < .001). The median times to repeat delivery were 790 days in LARC users, 659 days in non-LARC users, and 624 days among women who chose no contraceptive method. LARC use also delayed the time to subsequent delivery compared with no evidence of postpartum contraceptive use (HR, 0.34; P < .001). Compared with no postpartum contraceptive use, use of a non-LARC method postpartum had no effect on time to subsequent delivery (HR, 1.27; P = .24).
Fig. 2.
Time to and probability of subsequent delivery according to postpartum contraceptive method use or nonuse, among a cohort of previously delivered adolescent and young adult women (N = 4068), Denver, Colorado, 2011-2015. LARC delayed time to subsequent delivery compared with non-LARC, P <.001; LARC delayed time to subsequent delivery compared with no evidence of postpartum contraceptive use, P <.001; no postpartum contraceptive use and use of non-LARC methods postpartum showed no significant difference in time to subsequent delivery, P = .24. LARC, long-acting reversible contraceptive.
Discussion
In this retrospective analysis of adolescents and young women within one health system in the Denver metro area we found that postpartum contraceptive initiation was prevalent (45%), and repeat delivery within 33 months occurred in 16% of women. This rate is lower than previously published rates of repeat adolescent pregnancy, which have been reported in the 35%-42% range.14,15 Hispanic women in this cohort were more likely to initiate a method of postpartum contraception and more likely to have that method be a LARC device. Very young women (younger than age 15 years) were also more likely to initiate postpartum LARC use, which occurred most commonly across the entire cohort in the inpatient setting. Hispanic ethnicity and age were independently associated with postpartum LARC initiation, and did not interact with the primary outcome.
Postpartum contraception is advised by many national organizations concerned with improving maternal, perinatal, pregnancy, and early childhood outcomes, but target initiation rates have not been established.5,16 Compared with the 45% initiation rate in this cohort, in another large American cohort, almost two-thirds of women (63%) initiated postpartum contraception, but this was within a 6-month time frame and in a universal health care system.17 This analysis highlights a gap in guidelines regarding explicit goals for postpartum contraceptive initiation rates, which should serve as a quality indicator within and across health care systems.
Our analysis showed higher rates of postpartum LARC initiation from the inpatient setting. For women who did not initiate LARC during their inpatient admission, they were more likely to do so if they attended a postpartum visit. This suggests that for organizations or providers who wish to increase postpartum contraceptive uptake among adolescent and young adult women, offering inpatient LARC devices and ensuring postpartum follow-up might be effective. Accordingly, providers and other patient representatives should advocate for insurance coverage of LARC devices in the hospital and at the postpartum visit.18–20
This analysis also showed that Hispanic adolescents and young women were more likely to initiate postpartum contraception overall and LARC use, specifically, and that these behaviors were associated with reduced repeat pregnancy. In discussing these results with Denver Health advocates, they suggested these findings were the product of targeted interventions supported by the CFPI and favorable Title X funding, which we hypothesize might be one reason the very youngest of women initiated LARC use.7 Therefore, favorable policies and investments might have contributed to Denver Health’s success in delivering LARC devices to its adolescent and young adult patient population, which is primarily Hispanic. These data support the concept that a hospital-based, postpartum LARC campaign might improve pregnancy spacing and outcomes.
This retrospective, electronic health record data analysis was limited by several assumptions. Only patients who continued their care within the same health system were included in the assessment of repeat pregnancies and contraception; repeat pregnancies of women who delivered elsewhere or women who used postpartum contraception initiated at an alternate facility would have been misclassified. However, Denver Health serves as Denver’s primary safety net provider so many vulnerable patients return uniquely to that site of care. Additionally, gestational age calculations for pregnancies were not readily identifiable from unstructured electronic health record data; these would have allowed us to calculate interpregnancy interval and nonviable outcomes (eg, termination of pregnancy or spontaneous abortion) instead of using the time frame from delivery to delivery as our secondary outcome. The electronic record is also limited by the inability to determine which of the repeat pregnancies were planned vs unplanned and the time frame of and adherence to postpartum contraceptive use. The strengths of our study were the large sample size, the long and comprehensive time frame of observation and follow-up, and the ability to rapidly gather information from clinical and administrative data.
In conclusion, this analysis highlights a gap in our collective understanding of ideal postpartum contraceptive initiation rates. National advocacy and quality improvement organizations might want to consider if postpartum contraception guidelines should be issued with target quality metrics to assess program improvement over time. Our findings also suggest that contraceptive initiation rates might be enhanced with inpatient LARC campaigns, because this contraception method, received in the hospital setting, resulted in higher acceptance among this patient cohort. It also supports previous findings that short-acting methods were less effective compared with long-acting methods at promoting pregnancy spacing. Compared with young women, very young adolescents (age younger than 15 years) had the highest LARC uptake rates, suggesting targeted health care system efforts focused on ensuring the most vulnerable adolescents have access to postpartum contraception might be a successful approach for other health care systems to emulate.
Acknowledgments
The authors thank all of the women and men involved in collection of the data analyzed in this work and all of the women who participated in the study—their health, wellbeing, and successful pregnancy outcomes are the motivation for performing this work.
Funding for this project provided by the Adult and Child Consortium for Health Outcomes Research and Delivery Science, The Colorado Health Foundation, United States Grant 7179, the Eunice Kennedy Shriver National Institutes of Child Health and Human Development Women’s Reproductive Health Research K12 award (5K12HD001271-18), and the Doris Duke Charitable Foundation, United States.
Appendix A. Billing Codes Used in Analysis
| Category | ICD-9 | ICD-10 |
|---|---|---|
| Antepartum, subsequent pregnancy | V220 | Z3403, Z3402, Z3400, Z3401 |
| V221 | Z3482, Z3492, Z3483, Z3480, Z3490, Z3493, Z3481, Z3491 | |
| V230 | O0903, O0900, O0901, O0902 | |
| V231 | O0913, O0910, O0912, O0911, O09A0, O09A1, O09A2, O09A3 | |
| V232 | O09291 | |
| V233 | O0943, O0940, O0941, O0942 | |
| V2341 | O09212, O09219, O09211, O09213 | |
| V2342 | O0910, O0911, O0913, O0912 | |
| V2349 | O09292, O09299, O09293, O09291 | |
| V235 | O09292, O09291, O09293, O09299 | |
| V237 | O0933, O0930, O0931, O0932 | |
| V2383 | O09613, O09612, O09619, O09611 | |
| V2384 | O09623, O09629, O09621, O09622 | |
| V2385 | O09811, O09812, O09819, O09813 | |
| V2386 | O09822, O09821, O09823, O09829 | |
| V2387 | O3680X3, O3680X2, O3680X9, O3680X5, O3680X4, O3680X0, O3680X1 | |
| V2389 | O09892, O0971, O0973, O09899, O09893, O0970, O0972, O09891 | |
| V239 | O0990, O0993, O0991, O0992, Z362 | |
| V280 | Z36, Z360 | |
| V281 | Z36, Z361 | |
| V282 | Z36, Z3689 | |
| V283 | Z36, Z363 | |
| V284 | Z36, Z364 | |
| V285 | Z36, Z365 | |
| V286 | Z36, Z3685 | |
| V2881 | Z36, Z3689 | |
| V2882 | Z36, Z3689 | |
| V2889 | Z36, Z3681, Z3682, Z3683, Z3684, Z3686, Z3687, Z3688, Z3689, Z368A | |
| V289 | Z36, Z369 | |
| Pregnancy outcome | V270 | Z370 |
| V271 | Z371 | |
| V272 | Z372 | |
| V273 | Z373 | |
| V274 | Z374 | |
| V275 | Z3754, Z3752, Z3750, Z3753, Z3759, Z3751 | |
| V276 | Z3763, Z3760, Z3762, Z3764, Z3769, Z3761 | |
| V277 | Z377 | |
| Postpartum | V240 | Z390 |
| V242 | Z392 | |
| Subsequent pregnancy | V7242 | Z3201 |
| V222 | Z333 | |
| Vaginal birth | 650 | |
| Cesarean birth | 66970 | |
| 66971 | ||
| Birth control: short-acting | V2501 | Z30011, Z30015, Z30013, Z30016, Z30012 |
| V2511 | Z30017 | |
| V252 | Z30014 | |
| V255 | Z30430, Z302 | |
| Birth control: other | V2502 | Z30019 |
| V2503 | Z30018 | |
| V253 | Z308 | |
| V258 | Z309 | |
| V259 | Z309 |
ICD-9, International Classification of Diseases, Ninth Revision; ICD-10, International Classification of Diseases, 10th Revision.
Footnotes
The authors indicate no conflicts of interest.
References
- 1.Amjad S, Macdonald I, Chambers T, et al. : Social determinants of health and adverse maternal and birth outcomes in adolescent pregnancies: a systematic review and meta-analysis. Paediatr Perinat Epidemiol 2019; 33:88. [DOI] [PubMed] [Google Scholar]
- 2.Stevens J, Lutz R, Osuagwu N: Brief interpregnancy interval: are 75% of adolescent mothers unaware of the prematurity risk? Am J Public Health 2018; 108(suppl 1):S11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Conde-Agudelo A, Rosas-Bermúdez A, Kafury-Goeta AC: Birth spacing and risk of adverse perinatal outcomes. JAMA 2006; 295:1809. [DOI] [PubMed] [Google Scholar]
- 4.World Health Organization: Report of a WHO technical consultation on birth spacing. Available: https://apps.who.int/iris/bitstream/handle/10665/69855/WHO_RHR_07.1_eng.pdf. Accessed July 23, 2019.
- 5.March of Dimes: Birth Spacing and Birth Outcomes Available: https://www.marchofdimes.org/MOD-Birth-Spacing-Factsheet-November-2015.pdf. Accessed July 23, 2019.
- 6.Cohen R, Sheeder J, Arango N, et al. : Twelve-month contraceptive continuation and repeat pregnancy among young mothers choosing postdelivery contraceptive implants or postplacental intrauterine devices. Contraception 2016; 93:178. [DOI] [PubMed] [Google Scholar]
- 7.Colorado.gov: Taking the Inintended out of Pregnancy: Colorado’s Success with Long-Acting Reversible Contraception. Available: https://www.colorado.gov/pacific/sites/default/files/PSD_TitleX3_CFPI-Report.pdf. Accessed July 23, 2019.
- 8.Dube N: Colorado’s Family Planning Initiative. Available: https://www.cga.ct.gov/2015/rpt/2015-R-0229.htm. Accessed July 23, 2019.
- 9.Harrison MS, White C, Sheeder J, et al. : Adolescent Interpregnancy interval in Colorado among women with a last live birth between 2004 and 2013. J Adolesc Health 2019; 65:289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Winner B, Peipert JF, Zhao Q, et al. : Effectiveness of long-acting reversible contraception. N Engl J Med 2012; 366:1998. [DOI] [PubMed] [Google Scholar]
- 11.Secura GM, Allsworth JE, Madden T, et al. : The Contraceptive CHOICE Project: reducing barriers to long-acting reversible contraception. Am J Obstet Gynecol 2010; 203:115, e111 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bacon E, Budney G, Bondy J, et al. : Developing a regional distributed data network for surveillance of chronic health conditions: the Colorado Health Observation Regional Data Service. J Public Health Manag Pract 2019; 255: 498–507 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.UCLA. Institute for Digital Research & Education: Statistical Consulting: How Can I Estimate Relative Risk in SAS Using Proc Genmod for Common Outcomes in Cohort Studies?. Available: https://stats.idre.ucla.edu/sas/faq/how-can-i-estimate-relative-risk-in-sas-using-proc-genmod-for-common-outcomes-in-cohort-studies/. Accessed October 17, 2019.
- 14.Raneri LG, Wiemann CM: Social ecological predictors of repeat adolescent pregnancy. Perspect Sex Reprod Health 2007; 39:39. [DOI] [PubMed] [Google Scholar]
- 15.Damle LF, Gohari AC, McEvoy AK, et al. : Early initiation of postpartum contraception: does it decrease rapid repeat pregnancy in adolescents? J Pediatr Adolesc Gynecol 2015; 28:57. [DOI] [PubMed] [Google Scholar]
- 16.American College of Obstetricians and Gynecologists, and Committee on Obstetric Practice: Committee opinion no. 670: immediate postpartum long-acting reversible contraception. Obstet Gynecol 2016; 128:e32. [DOI] [PubMed] [Google Scholar]
- 17.Brunson MR, Klein DA, Olsen CH, et al. : Postpartum contraception: initiation and effectiveness in a large universal healthcare system. Am J Obstet Gynecol 2017; 217:55, e1 [DOI] [PubMed] [Google Scholar]
- 18.Hathaway M, Torres L, Vollett-Krech J, et al. : Increasing LARC utilization. Clin Obstet Gynecol 2014; 57:718. [DOI] [PubMed] [Google Scholar]
- 19.Rodriguez MI, Evans M, Espey E: Advocating for immediate postpartum LARC: increasing access, improving outcomes, and decreasing cost. Contraception 2014; 90:468. [DOI] [PubMed] [Google Scholar]
- 20.Moniz MH, Spector-Bagdady K, Heisler M, et al. : Inpatient postpartum long-acting reversible contraception: care that promotes reproductive justice. Obstet Gynecol 2017; 130:783. [DOI] [PMC free article] [PubMed] [Google Scholar]