key: cord-0051621-yev6kuv4
authors: Witwer, Elizabeth; Jones, Rachel K.; Fuentes, Liza; Castle, S. Kate
title: Abortion service delivery in clinics by state policy climate in 2017
date: 2020-10-16
journal: Contracept X
DOI: 10.1016/j.conx.2020.100043
sha: e7c7610e919c587b5934797d356964185eaef7fc
doc_id: 51621
cord_uid: yev6kuv4

OBJECTIVE: To examine service delivery in clinics that provided abortions in 2017, including differences by abortion policy climate. STUDY DESIGN: Using data from the Guttmacher Institute's 2017 Abortion Provider Census, we examine amount charged for abortion care, pregnancy gestation at which abortions were offered, number of days per week that clinics provided abortions, and types of non-abortion services offered. Our analysis focuses on the 808 clinic facilities that provided 95% of abortions that year. Measures were calculated nationally and according to whether the clinic was in a state we categorized as hostile, middle-ground or supportive of abortion rights. RESULTS: In 2017, 64% of clinics offered abortion at 11 weeks pregnancy gestation, and 22% did so at 20 weeks gestation. Supportive states had a higher density of clinics that provide abortion for every measured gestation than hostile states. Clinics charged an average of $549 for a surgical abortion at 10 weeks and $551 for medication abortion. Some 46% of clinics in supportive states offered abortion care five or more days per week, compared to 29% in hostile states. Most clinics offered stand-alone contraception and family planning (87%) and gynecological care (85%), but the proportion of clinics that provided these services was higher in supportive states (93% and 90%) than in hostile states (75% and 73%). CONCLUSIONS: A substantial proportion of abortion facilities provide a range of other health care services. Aspects of service delivery, such as number of days abortions are provided, may vary according to abortion policy climate. Implications statement. Onerous policies in states hostile to abortion rights may inhibit some facilities from providing abortion more days per week, and if so, could further burden patients obtaining abortion care in these states.

Access to abortion is often measured by the distance one must travel to obtain abortion care [1] [2] [3] . However, other dimensions of access, such as cost, gestational parameters, and the frequency and schedule by which services are provided can also facilitate or inhibit use of abortion care [3] . For example, the cost of abortion is a well-documented barrier to obtaining services [4] [5] [6] . This is unsurprising given that 75% of abortion patients are poor or low-income and many are uninsured or have insurance that does not cover abortion [7] . The majority of abortion patients pay out of pocket for care [7] , and in 2014, clinics charged an average of $508 for first-trimester surgical abortions and $535 for medication abortions [8] . Pregnancy gestational parameters can also be barriers to care. While the majority of abortions occur in the first-trimester, approximately 12% occur later in pregnancy [9] . The majority of clinics offered second-trimester abortion services in 2014, but availability decreased steeply after 15 weeks [8] .

Some clinics only offer abortions a few days a week [10] [11] [12] . White et al. interviewed abortion patients traveling to an Alabama clinic that only provided abortions one day a week [13] . For some individuals, the state's mandatory waiting period and the clinic's adherence to in-person visits to share state-mandated information led to delays in care and additional costs.

However, little is known about this aspect of provision at clinics nationwide.

Research has demonstrated that some individuals would prefer to obtain abortion care in settings where they obtain other kinds of health services [14] [15] [16] [17] . In 2017, the majority of clinics providing abortion care were non-specialized, meaning more than half of patients were J o u r n a l P r e -p r o o f Journal Pre-proof seen for other health care services [1] . Most specialized abortion clinics provide post-abortion contraceptive services [18, 19] , and 23% of clinics providing abortions offer transgender care [20] . But little else is known about the types of non-abortion services offered by abortion facilities.

Finally, the state abortion policy climate may affect multiple dimensions of abortion access. In 2014, 57% of U.S. women of reproductive age resided in states considered hostile to abortion rights, yet these states combined had fewer clinics than supportive states and were less likely to offer both very early and later abortion care [8] .

We use data from all known U.S. clinics that provided abortions in 2017 to examine abortion cost, pregnancy gestations at which abortions were provided, number of days of provision per week, and availability of non-abortion services. We describe differences by whether states were characterized as hostile, middle-ground, or supportive of abortion rights.

In turn, we add context to the number and distribution of clinics to offer a more nuanced understanding of abortion access in the United States.

Data are from the Guttmacher Institute's 2017 Abortion Provider Census (APC). We obtained study approval from Guttmacher Institute's federally registered institutional review board. The methodology for the APC is detailed elsewhere [1] , but we briefly outline the methods here.

In January 2018, we surveyed all clinics, physicians' offices and hospitals (n= 2,227) known or believed to have provided abortions in 2016 or 2017. The survey administered to nonhospital facilities included questions on the number of abortions provided in 2016 and 2017, type of abortion services available (medication abortion, surgical abortion 1 , or both), minimum and maximum pregnancy gestation at which abortions were offered, and amount charged for abortion at 10 and 20 weeks gestation. We asked nonhospital facilities: "In a typical 1 The authors recognize that some publications are intentionally shifting the language used to describe abortion, including utilizing the term "procedural" instead of "surgical" abortion [21] . In this paper, we refer to "surgical" abortion in order to accurately reflect the language that was used on our survey instrument in 2018. week in 2017, how many days per week did this facility provide abortion procedures?" We also asked nonhospital facilities: "Please indicate which of the following types of care this facility offered in 2017." Response options included contraception or family planning (excluding postabortion contraception), gynecological care (including STI testing and treatment, and sexual health care), prenatal and/or obstetric care, transgender-specific health services, and general health care.

We collected abortion caseload data directly from 59% of facilities, including 85% of clinics. Because the majority (95%) of abortions were provided by clinics in 2017 [1] , our analyses focus on these facilities.

We acquired at least some information (including instances where information was obtained from websites) on pregnancy gestational parameters from 80% of clinics, number of days abortions were provided from 73%, cost of services from 70%, and types of non-abortion services offered from 69%. We constructed weights for these measures based on facility type and abortion caseload; our weighting assumes that nonresponding facilities resembled those that provided the relevant data on abortion caseload and type of facility.

We multiplied amount charged for first-trimester surgical and medication abortion by facility caseload to reflect the caseload-adjusted amount clinics charged for care; in turn, we report mean amount paid by individuals. (We outline our decision not to incorporate a cost of living adjustment in Appendix A.) Because abortions at 20 weeks gestation are less common, we report the median amount charged rather than a facility caseload-weighted average. We use analyses from the Guttmacher Institute to classify states as having abortion policy contexts that were either supportive, middle-ground, or hostile to abortion rights [22] . In 2017, 29 states were classified as hostile, nine states as middle-ground, and 12 states as supportive [23] ( Table 1 ). The District of Columbia is excluded from this framework, and our climate-specific analyses, because it is more comparable to a city than a state [22] .

We distinguish between two types of abortion-providing clinic facilities. Specialized abortion clinics ("specialized clinics") are nonhospital facilities in which half or more of patient visits were for abortion services. Non-specialized clinics are nonhospital facilities in which fewer than half of visits were for abortion.

We first examine the distribution of abortion facilities and number of abortions by facility type and policy climate. We then provide weighted estimates of gestational parameters, (unweighted) density of clinics that provide abortion at different gestations, average amount charged for abortion services, number of days per week of abortion provision, and types of non-abortion services offered at clinics. We examined all outcomes except gestational parameters by clinic type; gestational parameters were examined for all clinics to allow for more stable estimates as numbers are small at later gestations. We did not perform significance testing for this descriptive study.

Of the 1,587 health care facilities known to provide abortion care in 2017, the majority, 63%, were in states supportive of abortion rights and 26% were in hostile states ( Table 2) .

Clinics made up the majority of facilities providing abortion care in hostile states (61%) compared to 46% in supportive states, where a substantial minority of providers were hospitals (33%) and physicians' offices (21%). Still, there were fewer clinics in hostile states than in supportive ones (255 vs. 459). And while approximately the same proportion of total abortions in 2017 occurred in supportive states and hostile states (43% and 45%, respectively), 58% of women of reproductive age lived in hostile states [24] .

Despite the uneven distribution of provider types, clinics provided the majority of abortions in all three policy climates, ranging from 99% in hostile states to 92% in supportive ones. Specialized clinics provided 79% of abortions in hostile states, while in supportive states, abortions were most commonly provided by non-specialized clinics (50%).

Almost all clinics provided abortions between 6 and 9 weeks of pregnancy gestation (99-100%), and nearly half (49%) offered abortions at 4 weeks (Table 3) . While the majority of clinics provided abortion at 10 weeks gestation (95%), only 64% did at 11 weeks, reflecting that 30% of clinics only provided medication abortion [1] . Less than a quarter of all clinics offered abortions at 20 weeks gestation (22%).

Journal Pre-proof Pregnancy gestation parameters varied by policy context. Abortions at 4 weeks gestation were offered by a greater proportion of clinics in supportive (53%) and hostile states (48%) than in middle-ground states (32%). A higher proportion of clinics in hostile (33%) and middle-ground (26%) states provided abortion at 20 weeks compared to clinics in supportive states (13%). However, when examining the number of clinics that offer abortion by gestation for every 100,000 women of reproductive age, supportive states had a higher density of clinics that provide abortion for every measured gestation than hostile states.

In 2017, clinics charged an average of $549 for a first-trimester surgical abortion (Table   4 ). When adjusted for inflation, this was approximately $20 more than 2014 [8] . Nonspecialized clinics charged more than specialized clinics for first-trimester surgical abortion, $578 compared to $534. The average amount charged for this service was highest in middleground states ($592) and lowest in hostile states ($534). The average amount clinics charged for a medication abortion at 10 weeks was $551; the cost was $568 at non-specialized clinics and $541 at specialized clinics. Nationally, the median amount charged by clinics for an abortion at 20 weeks gestation was $1,670 (range $410-$5,386) (data not shown).

More than one-third of all clinics provided abortions five or more days a week (39%), and this proportion was slightly higher for specialized clinics (44%) than non-specialized ones (36%) ( Table 5 ). More than one quarter (28%) of non-specialized clinics provided abortion care one day per week or less; this was the case for 5% of specialized clinics.

In all three policy climates, less than 10% of specialized clinics provided abortions one day a week. This varied more for non-specialized clinics, where 16% in supportive states provided abortion one day per week compared to 61% in hostile states. Of non-specialized clinics, 46% in supportive states provided abortions five or more days a week compared to 8% in hostile states.

Journal Pre-proof

In 2017, 87% of clinics offered contraception or family planning and 85% offered gynecological care (Table 6 ). About a quarter (27%) offered general health care, 23% offered transgender-specific health services, and 18% offered prenatal or obstetric care. While 100% of non-specialized clinics offered stand-alone contraceptive services, 65% of specialized clinics did so. Indeed, non-specialized clinics were more likely to provide all types of non-abortion care compared to specialized clinics.

A higher proportion of clinics in supportive states (93%) provided contraception or family planning care than in hostile states (75%); figures were similar for gynecological care (90% and 73%). A comparable proportion of specialized clinics in hostile and supportive states offered contraceptive services (62% and 61%) and gynecological care (58% and 50%), and these services were provided by 99-100% of non-specialized clinics regardless of policy context. About a third of non-specialized clinics across all three policy climates offered transgender-specific health services (31-39%). A higher proportion of specialized clinics in hostile states (11%) provided these services than in supportive ones (6%).

This study suggests that a substantial proportion of abortion facilities provide a range of other health care services and that other aspects of service delivery, such as number of days abortions are provided, may vary according to abortion policy climate.

Unsurprisingly, specialized abortion clinics were more likely to provide abortion care on more days than non-specialized clinics, with little variation between specialized clinics in hostile and supportive states. Notably, the majority (61%) of non-specialized clinics in hostile states provided abortion only one day a week (or less), whereas a minority in supportive states (16%) did so. The number of days a clinic provides abortion care is contingent upon numerous factors, including clinician availability, facility capacity, and level of patient need; our survey did not distinguish between clinics that only offered abortion on specified days each week, and those that reported the average number of days abortions were provided given patient demand.

However, research has documented that abortion facilities must expend considerable financial and human capital in order to comply with restrictions such as targeted regulations of abortion provider (TRAP) laws [12, 25, 26] , in-person visits for state-mandated counseling [27] , and other onerous administrative requirements [28] . These laws and regulations can require institutional and personnel adaptations that may divert financial resources and staff time away from providing care. Therefore, the higher proportion of non-specialized clinics in hostile states that only provided abortions one day per week could mean that restrictions inhibit some facilities from providing abortion more days per week. Health systems research documenting constraints to offering abortion care at the convenience and request of patients is needed.

While over 90% of clinics in supportive states offered stand-alone contraceptive care and gynecological services, this figure was lower, 73-75%, in hostile states. Virtually all nonspecialized clinics offered these services regardless of policy climate, and the overall difference between supportive and hostile states is due to the fact that a higher proportion of clinics in hostile states specialize in abortion care. These findings suggest that specialized abortion clinics in hostile states-which make up the majority of abortion clinics in that context-might have difficulty sustaining operations if they are subject to laws or other circumstances that even temporarily disrupt abortion provision [29] . What remains unknown is if specialized clinics face barriers to offering additional stand-alone services to meet different patient needs. A study of administrators at independent abortion clinics found that navigating insurance agreements and reimbursement inhibits integration of contraceptive care at abortion clinics [30] ; it is likely that barriers to insurance reimbursement extends to other types of services, as well.

The proportion of clinics offering abortions at 20 weeks pregnancy gestation was higher in hostile states than supportive ones. This difference is due, in part, to the higher proportion of clinics in supportive states offering only medication abortion [8] . However, the higher proportion of clinics that offer abortion at 20 weeks in hostile compared to supportive states does not necessarily equate to greater availability of services to those seeking care. There was a higher density of clinics in supportive states than hostile providing abortion at every gestation measured, suggesting individuals needing later abortion services in hostile states have fewer J o u r n a l P r e -p r o o f Journal Pre-proof clinics where they can access care, and, in turn, may have to travel greater distances [31] or wait longer to obtain an appointment, incurring associated costs and burdens.

In 2017, clinics charged about $550 for both first-trimester surgical and medication abortions. In prior analyses, medication abortion cost approximately $20 more [8] ; now that medication abortion is firmly integrated into abortion care, the price of the two procedures may have equalized. Still, most abortion patients pay out of pocket for care [7] and given that 40% of individuals in the United States cannot afford a $400 emergency expense [32] , this cost is likely to impose a barrier for some patients.

This study has several limitations. We were unable to obtain data on some of the outcomes examined in this study from 19%-31% of clinics. If nonresponding facilities were substantially different on these aspects of service delivery, our findings may be inaccurate.

Although we tried to identify all abortion facilities, it is possible that some were not identified [1] . The number of days clinics provide abortion care can vary over the course of a year, and our survey could not capture any such changes to service-delivery over 2017 [12] . Finally, our data are from 2017, and it is possible our findings do not reflect the current realities of service provision. For instance, sexual and reproductive health service provision has changed since 2017 due to the domestic gag rule imposed by new Title X rules [33] . Moreover, at the time of publication, many aspects of abortion provision were rapidly changing in response to the COVID-19 global pandemic [34] , and our findings may not be current in the wake of this public health crisis. Continued monitoring of these measures is critical to assess changes in service delivery in this evolving landscape.

This study reinforces past findings that state policy climate may shape aspects of abortion service delivery. Onerous policies in hostile states may inhibit some facilities from providing abortion more days per week, and if so, could further burden patients obtaining abortion care. Health systems research is needed to understand if clinics face constraints to offering abortion provision schedules that meet patient needs, and to understand if clinics would like to expand the range health care services offered on site.

J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f Table 4 . Caseload-adjusted average charge and range for surgical abortion at 10 weeks pregnancy gestation and for medication abortion, nationally, by U.S. clinic type, and by policy climate, unadjusted for cost of living, 2017

Surgical abortion (US $) Table 5 . Weighted percent of U.S. abortion clinics by number of days per week providing abortion care, by clinic type and policy climate, 2017 Table 6 . Weighted percentage of U.S. abortion clinics offering non-abortion health services, nationally, by clinic type, and by policy climate, 2017

Abortion incidence and service availability in the United States

Predicted changes in abortion access and incidence in a post-Roe world

The concept of access: Definition and relationship to consumer satisfaction

At what cost? Payment for abortion care by U.S. women. Women's Health Issues

Out-of-pocket costs and insurance coverage for abortion in the United States

Barriers to abortion care and their consequences for patients traveling for services: qualitative findings from two states

Characteristics of U.S. abortion patients in 2014 and changes since

Differences in abortion service delivery in hostile, middleground, and supportive states in 2014

the United States. Guttmacher Institute

Abortion access is limited, but this traveling doctor is determined to provide care

Seven states have only one remaining abortion clinic. We talked to the people keeping them open

Care churn-Why keeping clinic doors open isn't enough to ensure access to abortion

Experiences accessing abortion care in Alabama among women traveling for services

Patient attitudes toward early abortion services in the family medicine clinic

Urban female patients' perceptions of the family medicine clinic as a site for abortion care

Women's preference for receiving abortion in primary care settings

I don't want to go anywhere else": patient experiences of abortion in family medicine

How commonly do US abortion clinics offer contraceptive services?

Perceived and insurance-related barriers to the provision of contraceptive services in U.S. abortion care settings

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

Notes on language: Why we stopped using "surgical abortion

Troubling trend: more states hostile to abortion rights as middle ground shrinks

Policy trends in the States

Vintage 2017 postcensal estimates of the resident population of the United States n

Regulating abortion: impact on patients and providers in Texas

TRAP laws and the invisible labor of US abortion providers

The experiences and adaptations of abortion providers practicing under a new TRAP law: a qualitative study

Abortion reporting: promoting public health, not politics

COVID-19 Abortion Bans and Their Implications for Public Health

Challenges and strategies for contraceptive care in independent abortion clinics in the United States

Distance traveled to obtain clinical abortion care in the United States and reasons for clinic choice

Board of Governors of the Federal Reserve System

Trump administration's domestic gag rule has slashed the Title X network's capacity by half

Commentary: No-test medication abortion: a sample protocol for increasing access during a pandemic and beyond. Contraception

Note: Numbers of abortions are rounded to the nearest 10 and percentages may not add to 100 because of rounding. National figures include District of Columbia; policy climate figures exclude District of Columbia. Sources: Number of providers and abortions

We thank our colleagues Megan Donovan, Joerg Dreweke, Elizabeth Nash, and Alicia VandeVusse of the Guttmacher Institute for providing feedback on earlier drafts of the paper.Funding support for this study was provided by an anonymous foundation.

We considered adjusting for differences using a county-level cost-of-living index (COLI). However, pricing for abortion does not appear to operate according to the same market dynamics as goods and services such as food, shelter and transportation. For example, in 2017 even after adjusting for clinic caseload, the mean amount a clinic charged for a first-trimester surgical abortion in Texas was $603 compared to $539 in New York. The overall COLI for Texas counties that had a clinic that provided abortion care was 105 compared to 119 in New York. (These figures mean that Texas counties had mid-management living costs 105% of the average for all US areas in the last quarter of 2017 and this figure was 119% for relevant New York counties.) Therefore, even prior to adjustment for differences in cost of living, abortion patients in New York state paid less than patients in Texas. Within the state of New York, abortion patients in New York City paid less for a first-trimester surgical abortion than in other parts of the state, $506 and $578 respectively. (The COLI figure for each area was 185 and 112.) In regards to the latter differences, it is possible that the concentration of abortion providing facilities in the New York City area resulted in lower prices compared to other areas of the state.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. J o u r n a l P r e -p r o o f