Background
Inpatient insertion of long-acting reversible contraception and immediate postpartum tubal ligation allow women to initiate highly effective contraception before hospital discharge.
Objective
We measured rates of intrauterine device and contraceptive implant initiation and tubal ligations performed during delivery hospitalizations from 2016 to 2018 from a representative sample of US hospital discharges.
Study Design
We used the 2016 to 2018 National Inpatient Sample database, a 20% sample of all community hospital discharges in the United States, to identify delivery hospitalizations with concomitant intrauterine device insertion, contraceptive implant insertion, or tubal ligation. We performed weighted multivariable logistic regression to examine associations between possible predictors (age, delivery mode, payer, geographic region, and year) and odds of long-acting reversible contraception and tubal ligation, and to compare characteristics of users.
Results
This sample included 2,216,638 discharges, representing 20% of 11,083,180 delivery hospitalizations across the United States. Intrauterine device insertion increased from 2.2 per 1000 deliveries (2016) to approximately 5.0 per 1000 deliveries (2018; P <.0001); implant insertion increased from 0.3 per 1000 deliveries (2016) to 2.5 per 1000 deliveries (2018; P <.0001); tubal ligation procedures decreased (64.2 to 62.1 per 1000 deliveries; P <.0001).
Women who underwent a cesarean delivery had higher odds of having a tubal ligation than those who had a vaginal delivery (adjusted odds ratio, 8.83; 95% confidence interval, 8.73–8.97). Women aged <25 years had 7 times higher odds of receiving long-acting reversible contraception than of receiving tubal ligation (adjusted odds ratio, 7.38; 95% confidence interval, 6.90–7.90). Women with public insurance had almost 5 times the odds of receiving long-acting reversible contraception compared with those with commercial insurance (adjusted odds ratio, 4.83; 95% confidence interval, 4.59–5.06).
Conclusion
Rates of long-acting reversible contraception insertion continue to increase while the rates of inpatient postpartum tubal ligations slowly decline. Variations in patient characteristics are associated with receiving long-acting reversible contraception or tubal ligation.
Introduction
Access to long-acting reversible contraception (LARC), including intrauterine devices (IUDs) and contraceptive implants in the immediate postpartum period, is increasing across the country. Availability of highly effective postpartum contraception is associated with decreases in rapid repeat pregnancy. , Rates of immediate postpartum LARC insertion have steadily increased in the past decade, but remain low compared with rates of tubal ligation. Rates of inpatient tubal ligation during the postpartum period are declining. Understanding whether population subgroups have differential uptake of LARC and postpartum tubal ligation and whether changes have occurred may help us to evaluate equitable care over time.
Why was this study conducted?
Rates of postpartum long-acting reversible contraception (LARC) and tubal ligation are changing. This study updates national trends for inpatient procedures regarding highly effective contraception.
Key findings
Rates of immediate postpartum placement of LARC are increasing while rates of immediate postpartum tubal ligations are slowly declining, although absolute estimates of tubal ligations still exceed those of LARC insertions. Certain patient and hospital characteristics are associated with LARC and tubal ligation uptake.
What does this add to what is known?
This updated study (2016–2018) shows that in a nationally representative cohort, inpatient LARC placement rates are increasing more quickly than inpatient tubal ligation rates are declining.
Our objective was to use the most recently available nationally representative data to estimate trends in LARC insertion and tubal ligation rates, identify modifiers, and compare characteristics of those who receive LARC with those who receive tubal ligation.
Materials and Methods
This retrospective cohort study used data from the National Inpatient Sample (NIS), the largest publicly available database of all-payer, inpatient discharges in the United States. This database is part of the Healthcare Cost and Utilization Project (HCUP), sponsored by the Agency for Healthcare Research and Quality. It draws data from every state that participates in HCUP and represents 97% of the US population. This included 46 states in 2016 and 47 states in 2017 and 2018. The NIS, which has generated national estimates of healthcare utilization since 1988, is a stratified sample of 20% of discharges from US hospitals.
We generated a data sample of delivery-related discharges from 2016 to 2018 by identifying records that had at least 1 diagnosis or procedure code for vaginal birth or cesarean delivery using the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). We defined LARC as receiving an IUD or a contraceptive implant, and tubal ligation as the removal of the fallopian tubes (the specific ICD-10-CM codes are included in the Supplemental Table). ICD-10-CM does not distinguish between partial removal of fallopian tubes and salpingectomies, and therefore we assumed that both were included in our analysis. We excluded records if they contained codes for hysterectomy (n=2189), codes for both tubal ligation and IUD (n=80), tubal ligation and implant (n=7), or IUD and implant (n=10). These exclusions accounted for <1% of delivery-related discharges. We included data on hospital region (Northeast, Midwest, South, West) and payer source (grouped as commercial or noncommercial insurance), which included having public insurance options (eg, Medicaid) or being uninsured. We modeled our methods closely to Moniz et al who analyzed inpatient LARC and tubal ligation rates from 2008 to 2013 using the NIS to compare trends over multiple years.
We conducted statistical analyses using SAS software (version 9.4; SAS Institute, Cary, NC). All analyses used Agency for Healthcare Research and Quality-specified discharge weights to obtain nationally representative estimates. We calculated a national weighted estimate for delivery hospitalizations, and rates (count per 1000 deliveries) of IUD insertion, implant insertions, and tubal ligations. We performed simple logistic regression with time as a predictor to assess trends across time. We calculated the rates of inpatient postpartum LARC insertion and tubal ligation across patient subgroups. In a multivariable logistic regression model with LARC, tubal ligation, or neither as the outcomes, we estimated adjusted odds ratios (aORs) and 95% confidence intervals (CIs). Separately, we used multivariable logistic regression models to compare characteristics between LARC and tubal ligation users. We controlled for patient age, delivery mode, region, payer, and year. We considered a P value of <.05 as statistically significant. This study consisted of deidentified national data and was deemed exempt from review by the institutional review board of Columbia University Irving Medical Center for human subject research.
Results
This sample included 2,216,638 unweighted delivery discharges, representing 11,083,180 delivery hospitalizations across the United States in the 2016 to 2018 National Inpatient Survey. An estimated 698,299 women underwent an immediate postpartum tubal ligation, whereas an estimated 55,775 received a LARC device between 2016 and 2018. The LARC insertion rate increased from 2.4 per 1000 deliveries in 2016 to 7.5 per 1000 deliveries in 2018. IUD insertion increased from 2.2 per 1000 deliveries (2016) to approximately 5.0 per 1000 deliveries (2018), and implant insertion increased from 0.3 per 1000 deliveries (2016) to 2.5 per 1000 deliveries (2018) ( Figure ). The tubal ligation rate declined slightly between these dates (64.2 per 1000 in 2016 vs 62.1 per 1000 in 2018).
LARC use was highest among those aged <25 years, those using noncommercial insurance, and those residing in the Northeast ( Table 1 ). Women who had a cesarean delivery were more likely to receive LARC than women who had a vaginal delivery (aOR, 1.53; 95% CI, 1.47–1.59). Women with public insurance had almost 5 times the odds of receiving LARC compared with those with commercial insurance (aOR, 4.83; 95% CI, 4.59–5.06).
| Variable | LARC insertion | Tubal ligation | ||||||
|---|---|---|---|---|---|---|---|---|
| Rate | SE | aOR a | 95% CI | Rate | SE | aOR a | 95% CI | |
| Age, y | ||||||||
| <25 | 7.1 | 0.5 | 1.1 | 1.1–1.2 | 19.5 | 0.3 | 0.1 | 0.1–0.2 |
| 25–29 | 5.0 | 0.3 | 1.0 | 0.9–1.1 | 55.1 | 0.7 | 0.4 | 0.4–0.5 |
| 30–34 | 3.9 | 0.3 | 1.0 | 0.9–1.1 | 73.4 | 1.0 | 0.6 | 0.6–0.6 |
| ≥35 | 3.9 | 0.3 | 1.0 | reference | 121.3 | 1.8 | 1.0 | reference |
| Delivery mode | ||||||||
| Vaginal | 4.7 | 0.3 | 1.0 | reference | 18.9 | 0.4 | 1.0 | reference |
| Cesarean | 5.9 | 0.3 | 1.5 | 1.5–1.6 | 155.8 | 2.0 | 8.9 | 8.7–9.0 |
| Primary payer | ||||||||
| Private | 1.8 | 0.1 | 1.0 | reference | 52.4 | 0.8 | 1.0 | reference |
| Nonprivate | 8.4 | 0.5 | 4.8 | 4.6–5.1 | 74.1 | 1.1 | 2.1 | 2.0–2.1 |
| Geographic region | ||||||||
| Northeast | 7.7 | 1.0 | 1.7 | 1.6–1.8 | 50.5 | 1.6 | 0.8 | 0.8–0.8 |
| Midwest | 5.3 | 0.7 | 1.2 | 1.1–1.3 | 56.8 | 1.4 | 1.2 | 1.2–1.2 |
| South | 3.9 | 0.5 | 0.7 | 0.7–0.8 | 76.0 | 1.6 | 1.4 | 1.4–1.4 |
| West | 4.9 | 0.5 | 1.0 | reference | 55.6 | 1.4 | 1.0 | reference |
| Year | ||||||||
| 2016 | 2.4 | 0.3 | 1.0 | reference | 64.2 | 1.2 | 1.0 | reference |
| 2017 | 5.3 | 0.5 | 2.2 | 2.0–2.3 | 62.7 | 1.2 | 1.0 | 1.0–1.0 |
| 2018 | 7.5 | 0.7 | 3.1 | 3.0–3.3 | 62.1 | 1.3 | 0.9 | 0.9–1.0 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
