Ectopic Pregnancy Risk in Users of Levonorgestrel-Releasing Intrauterine Systems With 52, 19.5, and 13.5 mg of Hormone.

| We used Danish national databases (eTable in Supplement1)toidentifynulliparouswomenwithoutprevi- ous ectopic pregnancy, abdominal or pelvic surgery, infertility treatment, endometriosis, or prior use of a levonorgestrel IUS. They were followed up from January 1, 2001, or their 15th birthday, whichever came last, until July 1, 2021, age 35 years, pregnancy, death, emigration, or the occurrence of any exclusion criterion, whichever came first. Exposure to the levonorgestrel IUSs and other hormonal contraceptives (such as combined oral contraceptives) was modeled as a time-varying variable; women could con- tribute to more than 1 exposure group. Time exposed was defined from date of redeemed prescription until end of supply, defined as the end of the maximum approved duration of use for the levonorgestrel IUS, ie, 5 years for 52 mg and 19.5 mg and 3 years for 13.5 mg. All treat- ment periods were extended by 28 days due to possible ini- tiation delays. The outcome was a discharge diagnosis of ectopic preg- nancy as defined in the hospital register. The eTable in Supplement1 providesinformationondatadefinitions. Absolute rates were age and education standardized to the entire cohort. Age-, education-, and calendar year– adjusted relative rate ratios were calculated from Poisson regression models. Each levonorgestrel IUS dose was com- pared with use of other hormonal contraceptives for the maximum approved duration of use as well as by use for 1, 2, or 3 years. Also, the 3 doses were compared with each other. R software version 4.2 (R Core Team) was used. Two-sided 95% CIs that did not cross 1 defined statistical significance. The Danish Health Data Board and Danish Data Protec- tionAgencyapprovedthestudy.Register-basedstudiesarenot subjecttoethicsapprovalanddonotrequirepatientconsent inDenmark.Allaccessibledatawereanonymized.

| There were 963964 women followed up for 7.8 mil- lion person-years. For users of levonorgestrel IUS doses of 52 mg, 19.5 mg, and 13.5 mg and other hormonal contracep- tives, the median ages were 24, 22, 22, and 21 years, respec- tively, while the proportions with a university education were 33%, 25%, 29%, and 16%. There were 2925 ectopic pregnancies observed (including 35 with 52-mg, 32 with 19.5-mg, and 80 with 13.5-mg levonorgestrel IUS and 763 with other hormonal contraception). The adjusted absolute rate (per 10000 person-years) of ectopic pregnancy was 2.4 with other hormonal contracep- tives, 7.7 with 52-mg levonorgestrel IUS (difference, 5.3 [95% CI, 1.9-8.7]), 7.1 with 19.5-mg levonorgestrel IUS (dif- ference, 4.8 [95% CI, 1.5-8.0]), and 15.7 with 13.5-mg levonorgestrel IUS (difference, 13.4 [95% CI, 8.8-18.1]) (Table). Corresponding adjusted relative rate ratios were 3.4 (95% CI, 2.4-4.8), 4.1 (95% CI, 2.9-5.9), and 7.9 (95% CI, 6.3- 10.1). For each levonorgestrel IUS dose, the ectopic preg- nancy rate increased with duration of use (Table). The adjusted rate difference (per 10000 person-years) of ectopic pregnancy between the 19.5-mg and 52-mg levonorgestrel IUSs was −0.6 (95% CI, −3.8 to 2.6), with an adjusted rate ratio of 1.2 (95% CI, 0.7-1.9), and between the 13.5-mg and 52-mg levonorgestrel IUSs was 8.0 (95% CI, 3.3- 12.7), with a rate ratio of 2.3 (95% CI, 1.6-3.5) (Table). The rate difference between the 13.5-mg and 19.5-mg levonor- gestrelIUSswas8.6(95%CI,3.9-13.3),witharateratioof1.9 (95% CI, 1.3-2.9) (Table).

| Use of any levonorgestrel IUS was associated with a significantly increased risk of ectopic pregnancy compared with other hormonal contraceptives. The risk was highestwiththe13.5-mglevonorgestrelIUS,includingwhen compared with the other doses.

of this study included possible residual con- founding, although several confounding factors were con- trolled for by design and statistical analysis. Furthermore, the 13.5-mg and 19.5-mg levonorgestrel IUSs are exclusively approved for contraception and used interchangeably in nulliparous women, limiting the possibility of confounding by indication. Thefindingssuggestthatthe19.5-mglevonorgestrelIUS shouldbepreferredoverthe13.5-mglevonorgestrelIUS. Related articles pages 910 and 933 Supplemental content jama.com (Reprinted) JAMA March 21, 2023 Volume 329, Number 11 935 © 2023 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ by a Copenhagen University Library User on 10/20/2023 Table. Adjusted Incidence Rates, Rate Differences, and Rate Ratios of Ectopic Pregnancy According to Type of Levonorgestrel-Releasing Intrauterine System (IUS)a Type of hormonal contraception No. No. of events/10 000 person-years (95% CI) Adjusted incidence rate ratio according to duration of levonorgestrel IUS use (95% CI) b Womenc Person-yearsd Ectopic pregnancy events Age- and education-standardized incidence rate Age- and education-standardized incidence rate difference Maximum approved duration of use e 1y 2y 3y Comparison between levonorgestrel IUSs and other hormonal contraceptivesf Levonorgestrel IUS 13.5 mg 18 463 40 345 80 15.7 (11.4-21.2) 13.4 (8.8 to 18.1) 7.9 (6.3-10.1) 5.6 (3.7-8.4) 8.0 (6.2-10.5) 7.9 (6.3-10.0) 19.5 mg 25 594 33 812 32 7.1 (4.3-11.2) 4.8 (1.5 to 8.0) 4.1 (2.9-5.9) 2.3 (1.2-4.3) 3.7 (2.4-5.5) 4.1 (2.9-5.9) 52 mg 13 319 34 398 35 7.7 (4.7-11.9) 5.3 (1.9 to 8.7) 3.4 (2.4-4.8) 1.3 (0.5-3.4) 2.0 (1.1-3.5) 3.1 (2.1-4.7) Other hormonal contraceptivesf 650 742 3 394 019 763 2.4 (2.1-2.6) 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] Comparison between high-dose levonorgestrel IUS and low-dose levonorgestrel IUSs Levonorgestrel IUS 13.5 mg 18 463 40 345 80 15.7 (11.4-21.2) 8.0 (3.3 to 12.7) 2.3 (1.6-3.5) 4.4 (1.5-12.7) 4.1 (2.2-7.9) 2.5 (1.6-4.0) 19.5 mg 25 594 33 812 32 7.1 (4.3-11.2) −0.6 (−3.8 to 2.6) 1.2 (0.7-1.9) 1.8 (0.6-5.8) 1.9 (0.9-3.9) 1.3 (0.8-2.2) 52 mg 13 319 34 398 35 7.7 (4.7-11.9) 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] Comparison between the 2 low-dose levonorgestrel IUSs Levonorgestrel IUS 13.5 mg 18 463 40 345 80 15.7 (11.4-21.2) 8.6 (3.9 to 13.3) 1.9 (1.3-2.9) 2.4 (1.2-5.1) 2.2 (1.4-3.5) 1.9 (1.3-2.9) 19.5 mg 25 594 33812 32 7.1 (4.3-11.2) 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] 1 [Reference] a Information on time without hormonal contraception use not shown. b Adjusted for age and calendar year in 5-year intervals and highest educational level (elementary school only, secondary school only, skilled worker, academic education [ie, bachelor’s, master’s, or PhD degree]). c A total of 3187 women were exposed to 2 different doses of levonorgestrel IUS during follow-up, while 33 women were exposed to all 3 throughout their follow-up period. d Median number of years of use was 2.3 years (IQR, 0.8-3.0) for the 13.5-mg dose, 1.0 year (IQR, 0.4-2.1) for the 19.5-mg dose, and 2.2 years (IQR, 0.5-4.3) for the 52-mg dose. e Three years for the 13.5-mg levonorgestrel IUS and 5 years for the 19.5-mg and 52-mg levonorgestrel IUS. f Of all person-time with use of other hormonal contraceptives than levonorgestrel IUS, 94% accounted for use of combined oral contraceptives, 4% for use of progestin-only pills, 1% for combined vaginal ring, 0.5% for combined patch, and 0.5% for progestin-only injection. Letters 936 JAMA March 21, 2023 Volume 329, Number 11 (Reprinted) jama.com © 2023 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ by a Copenhagen University Library User on 10/20/2023 Amani Meaidi, MD, PhD Christian Torp-Pedersen, MD, MMSc Øjvind Lidegaard, MD, DMSc Lina Steinrud Mørch, PhD Author Affiliations: The Danish Cancer Society Research Center, Cancer Surveillance and Pharmacoepidemiology, Copenhagen, Denmark (Meaidi, Mørch); Department of Cardiology, North Zealand Hospital, Hilleroed, Denmark (Torp-Pedersen); Department of Gynecology, Rigshospitalet, Copenhagen, Denmark (Lidegaard). Accepted for Publication: January 11, 2023. Corresponding Author: Amani Meaidi, MD, PhD, The Danish Cancer Society Research Center, Cancer Surveillance and Pharmacoepidemiology, Strandboulevarden 49, 2100 Copenhagen Ø, Denmark ([email protected]). Author Contributions: Dr Meaidi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Meaidi, Torp-Pedersen, Mørch. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: Meaidi. Critical revision of the manuscript for important intellectual content:All authors. Statistical analysis: Meaidi, Torp-Pedersen. Administrative, technical, or material support: Lidegaard. Supervision: All authors. Conflict of Interest Disclosures: Dr Torp-Pedersen reported receiving grants from Bayer and Novo Nordisk. Dr Lidegaard reported receiving grants from Exeltis to conduct a phase 4 safety study on a new progestogen-only pill, on request from the European Medicines Agency. No other disclosures were reported. Data Sharing Statement: All data were deidentified, linked, and accessed through a secure server at Statistics Denmark. The authors do not have permission to share the data. See Supplement 2. 1. Gemzell-Danielsson K, Schellschmidt I, Apter D. A randomized, phase II study describing the efficacy, bleeding profile, and safety of two low-dose levonorgestrel-releasing intrauterine contraceptive systems and Mirena.Fertil Steril. 2012;97(3):616-622. doi:10.1016/j.fertnstert.2011.12.003 2. Nelson A, Apter D, Hauck B, et al. Two low-dose levonorgestrel intrauterine contraceptive systems: a randomized controlled trial.Obstet Gynecol. 2013;122 (6):1205-1213. doi:10.1097/AOG.0000000000000019 3. Kopp-Kallner H, Linder M, Cesta CE, Segovia Chacón S, Kieler H, Graner S.

of hormonal contraception and protective effects against ectopic pregnancy. Obstet Gynecol. 2022;139(5):764-770. doi:10.1097/AOG. 0000000000004726 4. Elgemark K, Graner S, McTaggart J, et al. The 13.5-mg, 19.5-mg, and 52-mg levonorgestrel-releasing intrauterine systems and risk of ectopic pregnancy. Obstet Gynecol. 2022;140(2):227-233. doi:10.1097/AOG.0000000000004846 Prevalence and County-Level Distribution of Births in Catholic Hospitals in the US in 2020 CatholichospitalscompriseagrowingsectoroftheUShealth care marketplace, with the number of them increasing 28% from 2001 to 2020. 1 Hospital adherence to theEthical and Religious Directives for Catholic Health Care Serviceslimitspeo- ple’s access to contraception during their delivery hospi- talizationandstandardtreat- mentofobstetriccomplications. 2 Thisisincreasinglyimpor- tant given evolving state-level threats to comprehensive reproductivehealthcare,suchasabortionbans.Thisstudyde- scribed the volume and distribution of births in US hospitals with Catholic affiliations and quantified county-level pat- ternsofCatholicandnon-Catholichospitalbirths.

|UsingAmericanHospitalAssociation(AHA)survey data,weidentifiedhospitalsthatreportedatleast10birthsin 2020,alongwithservice-relatedcharacteristics.Wethenclas- sifiedhospitalsasCatholicusing2019datafromCommunity CatalystandpubliclyavailabledatafromtheCatholicHospi- talAssociationasofSeptember2022. 1,3 We summed the total number of births per county from hospitalswithatleast10birthsthatprovideddatatotheAHA andcalculatedtheproportionofbirthsineachcountyoccur- ringinCatholichospitalsusingSASversion9.4(SASInstitute Inc).Forcountiesinwhich100%ofbirthsoccurredinaCatholic hospital, we assessed availability of alternative birthing op- tionsbycalculatingwhetheracontiguouscountyhadanon- Catholichospitalreportingatleast10births.WeusedArcGIS ProtocreatemapsdetailingtheproportionofbirthsinCatholic hospitals and the proportion of births overlaid with the total numberofbirthsbycounty.TheDukeHealthInstitutionalRe- viewBoarddeemedthisstudyexempt.

| Of the 3143 total counties in the US, 2491 were rep- resented by 6163 total hospitals reporting data to the AHA. A total of 2832 hospitals with at least 10 deliveries were iden- tified in 1618 counties in 2020, comprising 3563660 births. Of eligible hospitals, 421 (14.9%) were Catholic and covered 541626 births (15.2%). Catholic hospitals, compared with non-Catholichospitals,hadlargermediannumbersofannual births(867vs713),obstetricbeds(19vs16),andtotalhospital beds (206 vs 145) (Table). Although a smaller proportion of Catholic hospitals were classified by the Centers for Medicare & Medicaid Services as critical access hospitals compared with non-Catholic hospitals, a larger proportion were consid- ered rural referral centers. Overall,181counties(11.2%)hadadominantCatholicmar- ketshare,withatleast50%ofdeliveriesinthecountyoccur- ringinCatholichospitals( Figure).Thisincluded117counties (7.2%) for which 100% of births occurred in Catholic hospi- tals.Ofthese117counties,113(96.6%)wereadjacenttoatleast 1countythathadnohospitalwithatleast10deliveriesand11 (9.4%) were not adjacent to any county with a non-Catholic hospitalwithatleast10deliveries.TheFigurealsoshowsthe distribution of total births with the percentage of births in Catholichospitals.

| To our knowledge, this is the first analysis to de- scribedistributionofbirthsatCatholichospitalsatthecounty level in the US. More than 500000 US births occurred in Supplemental content Table. Characteristics of Catholic and Non-Catholic Hospitals With at Least 10 Deliveries in the US, 2020 Characteristics Catholic hospitals (n = 421) Non-Catholic hospitals (n = 2411) Births, median (IQR) 867 (445-1715) 713 (327-1630) Obstetric care beds, median (IQR) 19 (10-30) 16 (7-28) Total hospital beds, median (IQR) 206 (100-341) 145 (57-288) Critical access hospital, No. (%) 32 (7.60) 416 (17.25) Rural referral center, No. (%) 102 (24.23) 379 (15.72) Sole community provider, No. (%) 25 (5.94) 211 (8.75) Letters jama.com (Reprinted) JAMA March 21, 2023 Volume 329, Number 11 937 © 2023 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ by a Copenhagen University Library User on 10/20/2023