Transgender abortion patients and the provision of transgender-specific care at non-hospital facilities that provide abortions

Abstract

Objective

To estimate the number of transgender and gender non-binary (TGNB) individuals who obtained abortions in the United States and the extent to which abortion facilities offer transgender-specific health services.

Study design

We collected survey data from all known health care facilities that provided abortions in 2017. For the first time, the questionnaire included items about TGNB abortion patients and services.

Results

We estimate that 462 to 530 TGNB individuals obtained abortions in 2017 and that 23% of clinics provide transgender-specific care.

Conclusion

Several hundred abortions were provided to TGNB individuals in 2017, primarily at facilities that did not provide transgender-specific health services.

Implications

Findings from this study support efforts to implement and expand gender-inclusive and affirming care at health care facilities that provide abortion.

1. Introduction

Inclusive public health practices, policies and research that address the relationship between sex assigned at birth, gender identity and social determinants of health are integral to achieving health equity [1]. Anecdotal evidence suggests that abortion-providing clinicians and advocates have long recognized the need for high-quality, gender-affirming care for their patients, though little is known about these services at facilities that provide abortions. In this study, we generate national-level estimates of the number of abortions provided to transgender and gender non-binary (TGNB) patients in 2017 as well as the proportion of nonhospital clinic facilities that were providing transgender-specific health services.

2. Study design

Data for this study are from the Guttmacher Institute's 2017 Abortion Provider Census (APC). The APC has surveyed all known abortion-providing facilities in the United States since 1973 and provides the most comprehensive accounting of abortion incidence and clinical abortion service provision in the United States. A detailed study methodology has been documented extensively in other sources [2]; we briefly describe it here.

We mailed paper questionnaires to all suspected or known abortion providing facilities in January 2018 and collected data via mail, telephone, fax and email through March 2019. Questionnaires included items about abortion service provision, including the total number of abortions provided in 2016 and 2017. For the first time, we asked clinics and physician's offices, “In 2017, approximately how many abortions at this facility were provided to individuals identifying as transgender, gender non-binary, or gender non-conforming?” (We did not provide any definitions of these terms on the questionnaire.) We also asked nonhospital facilities: “Please indicate which of the following types of care this facility offered in 2017.” We included a list of non-abortion health services, including “Transgender-specific health services, including hormone therapy and gender-affirming primary care.” This option was intentionally written to be broad in an attempt to capture the different types of services that are considered “transgender-specific” care, as distinct from “transgender-friendly” care; “transgender-specific” was not further defined. While terminology is evolving to include language that is more reflective of peoples' lived experience, including terms such as “genderqueer” and “gender-expansive,” for conciseness we use the terminology “transgender and gender non-binary” (TGNB) when reporting results.

Fifty-nine percent of all facilities provided information about number of abortions. We used health department data to determine abortion caseloads for 19% of facilities, and made estimates for an additional 20% of facilities. The vast majority of abortions we counted in 2017 — 89% — came directly from abortion providers. While we received abortion counts directly from the majority of the facilities we surveyed, data on other measures were sometimes incomplete. We received information on the number of TGNB individuals who obtained abortions in 2017 from 597 (56%) of clinics and physician's offices (though only 484 provided a numeric response, see below), and 605 (57%) provided information about provision of transgender-specific health services. As with other questions on the survey, physicians' offices had lower response rates on the items pertaining to TGNB abortion patients and services (32% and 33%, respectively).

In order to generate national estimates, we constructed weights using information about facility abortion caseload and facility type (e.g., abortion clinic vs physician's office). On the original item asking for total number of TGNB individuals who obtained abortions in 2017, 57 facilities (10% of those that answered the item) wrote in some variation of “not applicable/na” and 113 (19%) clearly indicated “don't know.” (Some 35 facilities that indicated “don't know” or “na” provided trans-specific services.) Follow up calls with several facilities indicating “not applicable” suggested that this information was not routinely captured in their patient files and could not be answered with certainty. We used two approaches in constructing the relevant weight. The more conservative approach assumed a response of “not applicable” meant the facility had not provided abortions to any transgender patients, and the 57 facilities were recoded to “0.” The less conservative approach assumed a “not applicable” response was akin to “don't know” and categorized both groups as “missing.” Analyses using weights assume that non-responding facilities were similar to those that provided information on the item.

3. Results

In 2017, there were 1069 nonhospital facilities providing abortion care, and 85 (18% of those that answered the item) reported providing abortions to 230 TGNB individuals. The majority occurred in specialized or non-specialized clinics, though 2 (1%) were reported to be provided in physicians' offices. Our weighted estimates suggest that these figures represent between 462 and 530 TGNB abortion patients nationwide. TGNB patients accessed abortion in all four regions of the United States.

Some 122 clinic facilities provided transgender-specific health services, and our weighted estimate suggests that 23% of all U.S. clinics offering abortions provided this type of care. An additional 38 physicians' offices reported providing transgender-specific health care.

In the unweighted data, 58 (73%) nonhospital facilities that provided abortions to TGNB patients did not provide transgender-specific health care, and we estimate that only 70 of the 230 (30%) abortions provided to TGNB individuals occurred at facilities that provided transgender-specific health services.

4. Discussion

In 2017, non-hospital facilities provided several hundred abortions to TGNB patients. Some abortion-providing facilities do not routinely document the gender identity of abortion patients and for those that do, it may be underreported or not always captured due to concerns around stigma and confidentiality. One shortcoming of this analysis is that the estimated number of TGNB individuals obtaining abortions is less precise than other aspects of care assessed in the study; indeed, our estimate of abortions to TGNB individuals is likely lower than the actual incidence. While a substantial minority of abortion-providing clinics and physicians' offices offer transgender-specific health care, the majority of TGNB patients obtained their abortions at facilities that did not provide these services. The findings from this study validate continued efforts for health care facilities that provide abortions to provide, strengthen, and expand gender-inclusive and affirming care.