Abstract
This study assesses severe parental morbidity, cesarean deliveries, and preterm births among commercially and publicly insured trans people compared with cisgender people.
Transgender, nonbinary, and gender diverse (trans) people have worse health outcomes than cisgender individuals. Trans individuals can and do become pregnant,1 but there are limited data regarding pregnancy outcomes. Prenatal care access barriers,2 minority stress and stigma,3 and prior or ongoing use of testosterone may place trans people at heightened risk of perinatal complications. We evaluated pregnancy outcomes among commercially and publicly insured trans people.
Methods
We analyzed all deliveries from 2014 to 2018 from the Truven MarketScan Medicaid and commercial databases separately.4 This study was deemed exempt by the University of Michigan institutional review board.
Patients with male gender at the time of delivery (as listed in the data sets) or an International Classification of Diseases, Ninth Revision or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code consistent with gender dysphoria were identified as trans (eTable 1 in Supplement 1),5,6 following previous methodology with an 89.3% true-positive rate and 0.05% false-negative rate.6 Our primary outcome was severe parental morbidity (also known as severe maternal morbidity), a validated composite of indicators (eTable 2 in Supplement 2); secondary outcomes were cesarean delivery and preterm birth, outcomes not included in severe parental morbidity. We compared demographics, chronic conditions, and antenatal conditions (Table 1) between trans individuals and the remainder of the population using bivariate analysis. We used logistic regression to test the association between trans identification and outcomes, adjusting for age, race and ethnicity (for the Medicaid database), chronic conditions, and antenatal conditions. Analyses were conducted using SAS version 9.4 (SAS Institute Inc). Statistical significance was defined as a 2-sided P < .05.
Table 1. Characteristics of Publicly and Commercially Insured Trans and Cisgender Birthing Individuals, 2014-2018.
| Characteristics | Medicaid | Commercial insurance | ||||
|---|---|---|---|---|---|---|
| Cisgender individuals (n = 1 255 942) | Trans individuals (n = 256) | P valuea | Cisgender individuals (n = 1 465 565) | Trans individuals (n = 1651) | P valuea | |
| Age, mean (SD), y | 26.65 (5.44) | 23.48 (5.50) | <.001 | 30.22 (5.21) | 30 (5.87) | .16 |
| Age group, No. (%), y | ||||||
| 15-25 | 594 562 (47.3) | 176 (68.8) | <.001 | 285 748 (19.5) | 410 (24.8) | <.001 |
| 26-35 | 566 521 (45.1) | 75 (29.3) | 951 370 (64.9) | 943 (57.1) | ||
| ≥36 | 94 859 (7.6) | <11b | 228 447 (15.6) | 298 (18.1) | ||
| Gender, No. (%)c | ||||||
| Female | 1 255 942 (100) | 188 (73.4) | 1 465 565 (100) | 289 (17.5) | ||
| Male | 0 | 68 (26.6) | 0 | 1362 (82.5) | ||
| Race and ethnicity, No. (%)d | ||||||
| Black | 403 174 (32.1) | 103 (40.2) | <.001 | |||
| Hispanic | 72 809 (5.8) | 27 (10.6) | ||||
| White | 625 295 (49.8) | 86 (33.6) | ||||
| Other | 41 975 (3.3) | <11b | ||||
| Missing data | 112 689 (9.0) | 37 (14.5) | ||||
| Chronic conditions, No. (%) | ||||||
| Any condition | 187 273 (14.9) | 58 (22.7) | <.001 | 135 124 (9.2) | 134 (8.1) | .12 |
| Heart disease | 4975 (0.4) | <11b | 7766 (0.5) | <11b | ||
| Chronic hypertension | 55 930 (4.5) | <11b | 55 548 (3.8) | 38 (2.3) | .002 | |
| Liver disease | 757 (0.1) | <11b | 941 (0.1) | <11b | ||
| Kidney disease | 8555 (0.7) | <11b | 8454 (0.6) | <11b | ||
| Respiratory disease | 51 517 (4.1) | 15 (5.9) | .16 | 42 798 (2.9) | 61 (3.7) | .06 |
| HIV | 1242 (0.1) | <11b | 312 (0.02) | 0 | .55 | |
| Preexisting diabetes | 21 620 (1.7) | <11b | 20 830 (1.4) | 14 (0.9) | .05 | |
| Pulmonary hypertension | 389 (0.03) | <11b | 348 (0.02) | <11b | ||
| Sickle cell disease | 3174 (0.3) | <11b | 3277 (0.2) | <11b | ||
| Substance use disorders | 60 261 (4.8) | 28 (10.9) | <.001 | 6322 (0.4) | <11b | |
| Anxiety | 29 587 (2.4) | <11b | 31 622 (2.2) | 60 (3.6) | <.001 | |
| Depression | 34 833 (2.8) | 13 (5.1) | .03 | 26 713 (1.8) | 39 (2.4) | .10 |
| Anxiety or depression | 56 100 (4.5) | 16 (6.3) | .17 | 50 394 (3.4) | 90 (5.5) | <.001 |
| Prenatal conditions, No. (%) | ||||||
| Hypertensive disorders of pregnancy | 174 317 (13.9) | 28 (10.9) | .17 | 194 918 (13.3) | 152 (9.2) | <.001 |
| Multiple gestation | 20 723 (1.7) | <11b | 33 043 (2.3) | 57 (3.5) | .001 | |
| Intrauterine growth restriction | 15 167 (1.2) | <11b | 12 064 (0.8) | 11 (0.7) | .48 | |
| Gestational diabetes | 81 163 (6.5) | <11b | 109 452 (7.5) | 92 (5.6) | .003 | |
| Outcomes, No. (%) | ||||||
| Severe parental morbiditye | 9328 (0.7) | <11b | 9844 (0.7) | 16 (1.0) | .14 | |
| Cesarean delivery | 347 677 (27.7) | 54 (21.1) | .02 | 493 548 (33.7) | 362 (21.9) | <.001 |
| Preterm birth | 141 290 (11.3) | 33 (12.9) | .41 | 125 331 (8.6) | 133 (8.1) | .47 |
Statistical significance was defined as P < .05.
Percentages and P values could not be calculated due to low numbers of people.
Gender as listed in the MarketScan databases at the time of the delivery admission.
Race and ethnicity were ascertained based on self-report and separated in the database into 4 categories based on classifications provided in the Truven MarketScan Medicaid database: non-Hispanic Black, Hispanic, non-Hispanic White, and other (which included American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and unknown). A missing indicator was used for those missing race and ethnicity information. Race and ethnicity were included as covariates because they could have been potential confounders of the analysis and are known to be associated with a variety of obstetric and perinatal outcomes. These data are not available in the commercial data set.
Acute myocardial infarction, aneurysm, acute renal failure, adult respiratory distress syndrome, amniotic fluid embolism, cardiac arrest/ventricular fibrillation, conversion of cardiac rhythm, disseminated intravascular coagulation, eclampsia, heart failure/cardiac arrest during surgery or procedure, puerperal cerebrovascular disorders, pulmonary edema/acute heart failure, severe anesthesia complications, sepsis, shock, sickle cell disease with crisis, air and thrombotic embolism, blood products transfusion, hysterectomy, temporary tracheostomy, and ventilation. Due to rare prevalences, the following indicators may be combined for reporting purposes: (1) acute myocardial infarction and aneurysm; (2) cardiac arrest/ventricular fibrillation and conversion of cardiac rhythm; and (3) temporary tracheostomy and ventilation.
Results
We identified 256 trans people and 1 255 942 cisgender people in the Medicaid database and 1651 trans people and 1 465 565 cisgender people in the commercial database having had a delivery. Compared with cisgender people, trans people in the Medicaid database were younger (mean age, 26.65 [SD, 5.44] vs 23.48 [SD, 5.50] years), less likely to be White (49.8% vs 33.6%), and more likely to have a chronic condition (14.9% vs 22.7%) (all P < .001). Compared with cisgender people, trans people in the commercial data set had higher rates of some chronic conditions, such as anxiety or depression (3.4% vs 5.5%; P < .001).
Compared with cisgender people, trans individuals in the commercial database had similar rates of severe parental morbidity (0.7% vs 1.0%; P = .14); rates in the Medicaid database could not be calculated due to low numbers. Rates of cesarean delivery were significantly lower among trans people in both data sets (Medicaid: cisgender, 27.7%; trans, 21.1%; P = .02; commercial: cisgender, 33.7%; trans, 21.9%; P < .001). There was no significant difference in the rates of preterm birth in either data set (Medicaid: cisgender, 11.3%; trans, 12.9%; P = .41; commercial: cisgender, 8.6%; trans, 8.1%; P = .47) (Table 1).
In multivariable analysis, trans identification was not significantly associated with severe parental morbidity (Medicaid: adjusted odds ratio [OR], 1.28 [95% CI, 0.29-5.57]; P = .75; commercial: adjusted OR, 1.62 [95% CI, 0.98-2.69]; P = .06). Trans identification was associated with a lower odds of cesarean delivery (Medicaid: adjusted OR, 0.47 [95% CI, 0.25-0.88]; P = .02; commercial: adjusted OR, 0.55 [95% CI, 0.45-0.66]; P < .001) but not with the odds of preterm birth (Medicaid: adjusted OR, 1.37 [95% CI, 0.94-2.01]; P = .99; commercial: adjusted OR, 0.90 [95% CI, 0.75-1.09]; P = .28) (Table 2).
Table 2. Adjusted Odds Ratios for Severe Parental Morbidity, Cesarean Delivery, and Preterm Birth Among Publicly and Commercially Insured Trans and Cisgender Birthing Individuals, 2014-2018.
| Medicaid | Commercial insurance | |||
|---|---|---|---|---|
| Adjusted odds ratio (95% CI) | P valuea | Adjusted odds ratio (95% CI) | P valuea | |
| Odds of severe parental morbidity among trans vs cisgender people | 1.28 (0.29-5.57) | .75 | 1.62 (0.98-2.69) | .06 |
| Odds of cesarean delivery among trans vs cisgender people | 0.47 (0.25-0.88) | .02 | 0.55 (0.45-0.66) | <.001 |
| Odds of preterm birth among trans vs cisgender people | 1.37 (0.94-2.01) | .99 | 0.90 (0.75-1.09) | .28 |
Significant at P < .05.
Discussion
In this study using 2 national claims data sets, compared with cisgender people, identification as trans was not significantly associated with severe parental morbidity and preterm birth and was associated with lower rates of cesarean delivery. Difference in cesarean birth rates may result from differences in birth care clinicians or interventions. The similar outcomes occurred despite trans people having higher overall prevalence of chronic conditions. However, the prevalence of chronic conditions was lower than that reported in the general trans population, which may reflect the cross-sectional study design, differences between birthing and nonbirthing trans people, or undercapture in claims data.
Study limitations include that higher-risk subpopulations of trans people giving birth may have been missed, including those who had not identified as trans to their clinician. Additionally, the study may have been insufficiently powered to detect a difference in the rare outcome of severe parental morbidity, and testosterone use could not be evaluated.
Prospective data and larger sample sizes, as well as studies evaluating patient-reported outcomes, are needed to better characterize birth outcomes and for future quality improvement.
Section Editors: Jody W. Zylke, MD, Deputy Editor; Kristin Walter, MD, Senior Editor.
eTable 1. Details on Identifying Study Sample With Delivery (Vaginal and/or Cesarean)
eTable 2. Codes Used to Identify Severe Parental Morbidity (SPM)
Data Sharing Statement
References
- 1.Moseson H, Fix L, Hastings J, et al. Pregnancy intentions and outcomes among transgender, nonbinary, and gender-expansive people assigned female or intersex at birth in the United States: results from a national, quantitative survey. Int J Transgend Health. 2021;22(1-2):30-41. doi: 10.1080/26895269.2020.1841058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hoffkling A, Obedin-Maliver J, Sevelius J. From erasure to opportunity: a qualitative study of the experiences of transgender men around pregnancy and recommendations for providers. BMC Pregnancy Childbirth. 2017;17(suppl 2):332. doi: 10.1186/s12884-017-1491-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.White Hughto JM, Reisner SL, Pachankis JE. Transgender stigma and health: a critical review of stigma determinants, mechanisms, and interventions. Soc Sci Med. 2015;147:222-231. doi: 10.1016/j.socscimed.2015.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Adamson DM, Chang S, Hansen LG. Health Research Data for the Real World: The MarketScan Databases. Research and Pharmaceutical Division, Thomson Medstat; 2006. Accessed August 2, 2022. http://patientprivacyrights.org/wp-content/uploads/2011/06/Thomson-Medstat-white-paper.pdf
- 5.Proctor K, Haffer SC, Ewald E, Hodge C, James CV. Identifying the transgender population in the Medicare program. Transgend Health. 2016;1(1):250-265. doi: 10.1089/trgh.2016.0031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Blosnich JR, Cashy J, Gordon AJ, et al. Using clinician text notes in electronic medical record data to validate transgender-related diagnosis codes. J Am Med Inform Assoc. 2018;25(7):905-908. doi: 10.1093/jamia/ocy022 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Details on Identifying Study Sample With Delivery (Vaginal and/or Cesarean)
eTable 2. Codes Used to Identify Severe Parental Morbidity (SPM)
Data Sharing Statement