COVID-19 and Independent Abortion Providers: Findings from a Rapid-Response Survey.

CONTEXT: The ways in which the COVID-19 pandemic has affected abortion providers and abortion care, and the strategies clinics are adopting to navigate the pandemic, have not been well documented.
METHODS: In April-May 2020, representatives from 103 independent abortion clinics (i.e., those not affiliated with Planned Parenthood) completed a survey that included close-ended questions about how the pandemic, the public health response, and designations of abortion as a nonessential service had affected their clinic, as well as open-ended questions about the pandemic's impact. Analyses were primarily descriptive but included an exploration of regional variation.
RESULTS: All U.S. regions were represented in the sample. At 51% of clinics, clinicians or staff had been unable to work because of the pandemic or public health responses. Temporary closures were more common among clinics in the South (35%) and Midwest (21%) than in the Northeast and West (5% each). More than half of clinics had canceled or postponed nonabortion services (e.g., general gynecologic care); cancelation or postponement of abortion services was less common (25-38%, depending on type) and again especially prevalent in the South and Midwest. Respondents reported the pandemic had had numerous effects on their clinics, including disrupting their workforce, clinic flow and work practices; increasing expenses; and reducing revenues. State laws (including designations of abortion as nonessential) had exacerbated these difficulties.
CONCLUSIONS: Although independent abortion clinics have faced considerable challenges from the pandemic, most continued to provide abortion care. Despite this resiliency, additional support may be needed to ensure sustainability of these clinics.

Perspectives on Sexual and Reproductive Health, 2020, 52(4):TK, doi:10.1363/psrh.12163

On January 30, 2020, the World Health Organization officially declared COVID‐19 a public health emergency. Three months later, there had been half a million cases and more than 23,000 deaths worldwide; by November 2020, these numbers surpassed 50 million cases and close to 1.3 million deaths.2, 3 Government statistics, both globally and in the United States, continue to document the pandemic's direct effects on such outcomes as morbidity and mortality.3, 4 However, the pandemic can affect health in indirect ways, such as through broader disruptions to the health care system, including reduced availability of various types of primary and chronic disease care and changes in the health care workforce.5, 6, 7, 8, 9, 10, 11

Abortion providers likely have also been affected by the pandemic, though research to date has not examined their experiences. Investigating the events occurring in abortion clinics, in particular, is important for two key reasons. First, to be able to provide abortion care, clinics in some areas of the country must rely on doctors who travel from out of state;12, 13 therefore, the discouragement and attendant difficulties of travel during the pandemic could uniquely affect abortion clinics. Second, in addition to having to navigate the pandemic itself and related public health responses, abortion clinics in some states may have been affected by their governments’ designations of abortion as a nonessential service. These designations essentially banned abortion care in some states during the early weeks of the pandemic.14, 15 In particular, between March 17, 2020, and May 8, 2020, 12 states temporarily declared abortion a nonessential service. While most of these designations were no longer in effect by the end of April 2020,15, 16, 17 it is important to document providers’ experiences with them. To date, the scholarly literature about abortion during the COVID‐19 pandemic has focused primarily on the designation of abortion as a nonessential service or on clinical care changes designed to reduce risk of COVID‐19 transmission.12, 14, 18 In this study, however, we extended this research to examine experiences of clinics during the early months of the pandemic and to include both clinics that were subject to state‐mandated abortion service restrictions and those that were not, as well as to assess the pandemic's impact on the abortion care workforce.

To begin to document the ways that the pandemic and government responses to it have affected abortion clinics and their workforce, we conducted a rapid‐response survey during the early stages of the pandemic. More specifically, the survey was designed to document the ways in which the COVID‐19 pandemic, public health responses and the designations of abortion as a nonessential service affected abortion providers and abortion care, and to explore the strategies that clinics and staff adopted to navigate the pandemic.

METHODS

Sample and Recruitment

Between April 16, 2020, and May 22, 2020, we recruited independent abortion providers to participate in a brief (10–15 minutes) online survey about how COVID‐19 had affected their abortion clinic. For the purposes of this study, we defined independent abortion clinics as those not affiliated with Planned Parenthood. Close to 60% of abortions in the United States are provided by such clinics.

We recruited participants via emails to the Abortion Care Network (ACN) and Abortion Clinical Research Network electronic mailing lists, emails we sent directly to contacts at clinics that are part of the ACN and emails that colleagues sent to independent abortion clinics with which they had existing relationships. Approximately 110 independent clinics are members of the ACN, and 70 clinics are members of the Abortion Clinical Research Network. Although overlap exists between the two networks, 37 clinics that are members of the Abortion Clinical Research Network are not affiliated with Planned Parenthood or ACN, and are primarily affiliated with academic institutions or hospitals. We estimate that the email messages sent to the electronic mailing lists and the direct emails to contacts reached about 150–165 unique independent abortion clinics. Our unit of analysis was individual clinics, and thus we sought one respondent (preferably the clinic manager or director) from each clinic who could describe experiences at their site. We specifically targeted clinic managers and directors by addressing emails to them and by identifying them as our population of interest within consent documents; however, we accepted responses from any individual at an independent abortion clinic who was 18 or older, could complete the survey in English and could report on the experiences of clinic staff. The institutional review board at the University of California, San Francisco, provided ethical approval for the study.

Data Collection

We first asked potential respondents to review an online informed consent form before initiating the survey. Those who gave their electronic consent were then asked a series of close‐ended questions about their role at the clinic and about their clinic's characteristics (regional location, types of services and number of abortions provided, gestational limit and number of clinicians) in 2019. Next, we asked a series of questions about the experiences of clinicians who provided abortion care, another series about the experiences of nonclinician staff members, and a third about the experiences of patients since the onset of the pandemic. These included close‐ended questions regarding how the COVID‐19 pandemic and the public health responses to it may have affected the clinic (e.g., whether any clinicians or staff were unable to work because they were under quarantine, in a high‐risk group, sick with possible COVID‐19, unable to travel owing to public health restrictions, or needed to provide childcare or perform other caregiving responsibilities) and whether patients’ appointments had been canceled or postponed because of potential COVID‐19 exposure or travel restrictions. We also asked respondents to indicate whether their clinic had been forced to cancel or postpone any particular clinical services because of COVID‐19 (e.g., abortion services, gynecologic services or STI testing) and whether the clinic had closed for any period of time because of the pandemic or associated public health responses. Most of the questions also had open‐ended response options that allowed participants to report other experiences, so that we could ensure we were documenting the full range of the pandemic's effects on clinics. Respondents also reported whether their state had declared abortion to be an essential health care service.

Lastly, we asked respondents open‐ended questions that allowed them to describe any other ways in which COVID‐19 had affected their clinic and the services they were able to provide. The goal of these questions was to allow respondents to share any experiences that we did not directly ask about or to expand on any experiences that we did ask about. We also asked respondents whether they directed or oversaw more than one clinic; if they did, we invited them to complete a separate survey for each of the other clinics. We remunerated participants with a $20 gift card for their time.

Analysis

Our unit of analysis was the individual clinic. Analyses of responses to close‐ended questions were primarily descriptive, and we present the proportion of respondents reporting each of the experiences discussed in the survey. Because abortion laws, the availability of abortion care and the designation of abortion as a nonessential service during the pandemic all vary by region,15, 22 we used chi‐square tests (or, when cell sizes were small, Fisher's exact tests) to identify regional differences in clinic experiences. As a post hoc analysis, we also examined whether having had to cancel or postpone abortion services or temporarily close a clinic was associated with declarations of abortion as a nonessential service, again using Fisher's exact tests.

Because the response option for the question that asked whether the respondent's clinic had had to cancel or postpone specific services was a checkbox, we could not determine whether an unchecked box meant that the clinic had not had to cancel or postpone a service they had been providing, or whether the clinic simply had not provided that service in 2019 (which would have made the question irrelevant for that clinic). However, we had asked which abortion services clinics provided in 2019, and thus reran our analyses regarding postponement or cancelation of abortion services by restricting the denominator to the clinics that had provided that type of abortion care in 2019.

For open‐ended responses, we (S.R. and C.J.) reviewed the responses and created a list of codes based on themes that emerged in the data. S.R. then created a codebook, which S.R. and R.S. used to separately code open‐ended responses. S.R. and R.S. resolved any coding discrepancies through discussion. We coded for every theme present within an open‐ended response, and thus each response could fall into more than one coding category. As this was an exploratory study designed to identify the range of experiences among independent abortion clinics, we present the findings by theme and note whether a theme was relatively common or uncommon, but we do not indicate numerically how common each theme was.

RESULTS

Sample

We received 121 surveys during the study period. We omitted from the analysis one survey from a respondent who worked at a Planned Parenthood clinic, as well as 12 surveys from individuals whose responses we could not use because they failed to answer at least three questions about how the pandemic (and the public health responses to it) had affected their clinic. Lastly, we omitted five surveys because we had received a response from another person at the same clinic. In deciding which response to keep from the clinics with more than one survey, we prioritized the survey that was taken at the later date—on the assumption that it would likely capture a broader range of experiences—unless it had been completed by someone other than the clinic manager or director. Four clinic directors completed individual surveys for more than one clinic within their networks, which yielded 12 surveys, all of which we kept. Our final sample consisted of 103 clinics.

Respondents reported on experiences within clinics in all regions of the country; 21% of clinics were located in the Northeast, 25% in the Midwest, 31% in the South and 22% in the West (Table 1). In 2019, almost all clinics had provided medication abortion (98%) and first‐trimester aspiration abortion (95%), and more than three‐fourths had provided abortions in the second trimester or later (77%). Eighty percent of clinics had a gestational limit of at least 14 weeks, and 50% had a limit of at least 20 weeks. About one‐fourth of clinics had one or two clinicians who provided abortions in 2019, while more than half had four or more such clinicians. The modal number of medication abortions that participating clinics provided in 2019 was 100–499; the same was true for first‐trimester aspiration abortions and second‐trimester or later abortions. However, nearly 20% of clinics provided more than 1,500 medication abortions in 2019, and an even larger proportion provided that many first‐trimester aspiration abortions. Approximately half of respondents were clinic directors, managers, owners or other administrative/executive staff; about 40% described themselves as medical directors or clinicians; and 10% had other roles, such as working on policy or providing ultrasounds.

TABLE 1

Percentage of surveyed independent abortion clinics with selected characteristics, 2020

Characteristic	% (N=103)
Location
Northeast	21
Midwest	25
South	31
West	22
Abortion services provided in 2019
Medication abortion	98
First‐trimester aspiration abortion	95
Second‐trimester or later abortion	77
Gestational age limit in 2019
<14 weeks	20
≥14 weeks but <20 weeks	30
≥20 weeks	50
No. of clinicians in 2019
1	12
2	15
3	18
4	24
≥5	32
No. of medication abortions in 2019
0	2
1–99	8
100–499	35
500–999	29
1,000–1,499	8
≥1500	18
No. of first‐trimester abortions in 2019
0	5
1–99	4
100–499	28
500–999	21
1,000–1,499	17
≥1500	25
No. of second‐trimester or later abortions in 2019
0	22
1–99	15
100–499	40
500–999	14
1,000–1,499	5
≥1500	5
Respondent's role
Manager/director/owner/chief executive officer	53
Medical director/physician/advanced practice clinician	37
Other staff	10

Note: Percentages may not total 100 because of rounding.

Close‐Ended Responses

Half (51%) of clinics reported having had clinicians or staff members who were unable to work because of the pandemic or associated public health responses. About one in five clinics reported that one or more of their clinicians had been unable to provide care because they were quarantined (23%), in a high‐risk group (21%) or sick with a possible COVID‐19 infection (20%; Table 2). At nearly one in six clinics, clinicians had been unable to provide care because they were subject to COVID‐related travel restrictions (15%); others had been reassigned to meet COVID‐related responsibilities (13%) or had been unable to perform clinical duties because of childcare (12%) or other caregiving (5%) responsibilities. Among nonclinical staff, these proportions were generally even higher; the most common reason for being unable to work was childcare responsibilities (50%), followed by being sick with possible COVID‐19 (45%), being in quarantine (44%), belonging to a high‐risk group (33%) and having other caregiving responsibilities (18%).

TABLE 2

Percentage of clinics with selected experiences during the COVID‐19 pandemic, by region

Measure	Employee type	Northeast (n=22)	Midwest (n=26)	South (n=32)	West (n=23)	All (N=103)
CLINICIAN/STAFF‐RELATED
Reasons unable to work
Quarantined	Clinician	23	23	19	30	23
	Staff	50	38	50	39	44
Belonged to a high‐risk group	Clinician	18	12	25	30	21
	Staff	23	38	34	35	33
Sick with possible COVID‐19	Clinician	32	12	19	22	20
	Staff	55	46	34	48	45
Travel restrictions	Clinician	9	12	28	4	15†
	Staff	na	na	na	na	na
Had other COVID‐related responsibilities	Clinician	14	12	16	9	13
	Staff	na	na	na	na	na
Childcare responsibilities	Clinician	5	15	13	13	12
	Staff	50	38	59	52	50
Other caregiving responsibilities	Clinician	9	4	3	4	5
	Staff	18	15	31	4	18†
Other travel/transportation barriers‡	Clinician	0	4	3	0	2
	Staff	5	0	6	0	3
Legal barriers‡	Clinician	0	0	9	0	3†
	Staff	0	0	3	0	1
Other financial challenges‡	Clinician	0	4	3	0	2
	Staff	na	na	na	na	na
Fear of getting sick‡	Clinician	na	na	na	na	na
	Staff	5	8	6	4	6
Other staffing challenges‡	Clinician	na	na	na	na	na
	Staff	5	0	3	0	2
PATIENT‐RELATED
Reasons patients postponed/canceled appointments
COVID‐19 symptoms/exposure	na	32	46	53	35	43
Travel restrictions	na	9	12	19	9	13
Legal restrictions‡	na	0	0	22	4	8**
Logistical reasons‡	na	5	0	0	9	3
STATE‐RELATED
State declared abortion to be an essential service§						***
Yes	na	64	65	14	70	52
No	na	36	19	54	26	34
Explicitly declared that abortion is not an essential service	na	0	15	32	4	14

p < .01.

p < .001.

p < .10.

Survey did not specifically ask about this reason, but some respondents mentioned it as a write‐in response.

Four respondents did not answer the question about whether their state had declared abortion to be an essential service, but noted in open‐ended responses that their state did not explicitly designate abortion as essential; three of them, however, reported that they had been able to choose their own designation. We did not include these four responses in the analyses presented, but the findings regarding regional differences would not have changed if we had classified the respondents’ state as having designated abortion as essential.

Notes: Percentages may not total 100 because of rounding. p values denote regional differences. na=not applicable.

***

In addition, more than 40% of respondents reported that they had canceled or postponed appointments because patients had had COVID‐19 symptoms or had been exposed to the virus, and 13% had done so because patients had been subject to COVID‐related restrictions on travel. No regional differences were evident in the proportions of respondents reporting that their clinic had a clinician or staff person unable to work or that they had had to cancel or postpone patient appointments because of the pandemic or the general public health response.

Finally, half of the clinics were located in states that had declared abortion to be an essential health care service. In this case, however, regional differences were evident: More than 60% of respondents in the Northeast, Midwest and West reported that their state had declared abortion an essential service, compared with only 14% of those in the South.

Most clinics had had to cancel or postpone at least some clinical services. Most commonly, respondents reported disruptions to gynecologic services (59%), contraceptive visits (55%) and STI tests (45%; Table 3). Fewer clinics reported disruptions to abortion services; 38% had canceled or postponed first‐trimester aspiration abortions, 27% second‐trimester or later abortions and 25% medication abortions. While there were no regional differences in the cancelation or postponement of nonabortion services, the proportion of clinics that had canceled or postponed first‐trimester aspiration abortions was higher in the South (66%) and Midwest (38%) than in the Northeast (9%) and West (26%). Likewise, cancelation or postponement of second‐trimester or later abortions or of medication abortions was highest in the South and Midwest and lowest in the Northeast and West. Findings for cancelation and postponement of abortion services were similar in the sensitivity analysis in which we restricted the sample to clinics that confirmed having provided the relevant services in 2019 (not shown). Finally, 19% of respondents reported having had to close their clinic temporarily (Table 3); the proportion was lowest in the Northeast (5%) and West (5%) and highest in the Midwest (21%) and South (35%).

TABLE 3

Percentage of clinics that canceled or postponed selected services or that temporarily closed, by region

Measure	Northeast (n=22)	Midwest (n=26)	South (n=32)	West (n=23)	All (N=103)
Canceled/postponed service
Gynecologic services	73	54	50	65	59
STI tests	41	54	47	35	45
Contraceptive care	64	58	56	43	55
Medication abortion**	5	27	44	17	25
First‐trimester aspiration abortion***	9	38	66	26	38
Second‐trimester or later abortion***	5	19	59	13	27
Walk‐in services‡	0	19	13	0	9
Trans care‡	5	8	3	9	6
Clinic temporarily closed*	5	21	35	5	19

p<.05. **p<.01.

p<.001.

Survey did not specifically ask about this service, but some respondents mentioned it in a write‐in response.

Note: p values denote regional differences.

Cancelation or postponement of abortion services and temporary closure of clinics also varied according to whether states considered abortion an essential service. More than 70% of clinics in states that had explicitly declared abortion to be a nonessential service had canceled or postponed appointments for one or more types of abortion or had temporarily closed, compared with 10–20% of clinics in states that classified abortion as an essential service (Table 4).

TABLE 4

Percentage of clinics that canceled or postponed abortion appointments or had to close temporarily, according to whether they were in a state that declared abortion to be an essential service

Measure	State declared that abortion is an essential service
	Yes (n=51)	No (n=34)	Explicitly declared abortion is not essential (n=14)
Canceled/postponed appointment
Medication abortion**	20	18	71
First‐trimester aspiration abortion***	20	38	93
Second‐trimester or later abortion***	14	26	71
Closed temporarily ***	10	10	77

p<.01.

p<.001.

Open‐Ended Responses

The themes we identified from open‐ended responses largely echoed the topics we asked about in the closed‐ended questions and illustrate the workforce, service provision, and legal and community challenges that clinics faced while striving to provide services during the early months of the pandemic. However, additional ways in which these challenges may affect clinics over time were also identified.

Respondents described facing challenges related to maintaining and sustaining a workforce during the pandemic, and having to implement changes in work processes as a result. Many respondents mentioned having to hire, lay off or furlough staff. Some of the hiring stemmed from a need to implement safety protocols related to reducing COVID‐19 risk among staff and patients, such as spacing out appointments to minimize the number of people coming into contact with each other and taking the temperature of people about to enter the clinic. A few clinics had hired additional security guards to protect against the ongoing—and in some instances growing—protests at their clinic. Yet also common was the need for clinics to lay off or furlough staff, often because of declines in patient volume and in the ability to provide care. Themes related to reducing COVID‐19 risk, to legal and community context, and to changes in volume are described in more detail below.

Multiple respondents mentioned that their clinics had made substantial pandemic‐related changes to work processes in order to protect the health and safety of their staff and patients, such as splitting staff into two teams that worked asynchronous schedules, having staff work longer hours to ensure the spacing of appointments and providing as much care remotely as possible. For example, the deputy director of a Midwestern clinic reported that the clinic had split the staff into two teams:

“The first team works Monday through Wednesday, and the second team works Thursday through Saturday; in case a staff member on one team gets sick, then we can still have [the other] team available to continue services. We have also expanded hours; we are usually closed on Mondays, but are now seeing patients.”

Respondents noted that their clinics had had to address many challenges that staff were experiencing, including fear of getting COVID‐19, the need to orient to numerous changes in protocols during the first weeks of the pandemic, and—because of safety‐related changes in work processes—the absence of staff members’ usual sources of social and emotional support and emotional outlets. Respondents from clinics in states that had declared abortion to be a nonessential service also reported challenges navigating the uncertainty related to state laws and litigation.

To support their workforce, many clinics had taken such steps as offering hazard pay, supplying lunch for staff much more frequently than usual, providing opportunities for staff to talk collectively about their emotions and conveying their commitment to keeping everyone as safe as possible.

Respondents mentioned multiple challenges related to continuing to provide abortion services during the early months of the pandemic and outlined how their clinics had responded to them. They also indicated making changes to reduce COVID‐19 risk for staff and patients and having to navigate these new logistical challenges.

Steps that clinics had taken to reduce COVID‐19 risk included implementing and maintaining health protocols and reducing the likelihood of exposure. For example, the owner of a Midwestern clinic wrote that the pandemic had affected “every single aspect of how we schedule, how we see patients in the clinic, screening for any COVID symptoms, how we interact with each patient, for example we ask them to use hand sanitizer multiple times during their stay. We always kept a very clean clinic but this has now gone to a new level.” Other practices the clinic had instituted included basic public health measures such as taking patients’ and staff members’ temperatures at the door, screening people for COVID‐19 risk factors, and requiring patients and staff to wear masks.

Respondents also reported multiple changes to clinic flow, such as limiting the number of support people within the clinic, requesting that patients wait in their car to minimize their time in the waiting room, spacing out chairs in the waiting room, reducing the number of patients to facilitate sanitization between appointments and reducing contact between patients. A few respondents mentioned that their clinic was supplying meals to staff to minimize personal interactions and hence reduce risk of exposure. Clinics had also adapted new clinical practices, such as shifting to medication abortion whenever feasible. Respondents also commonly described providing care remotely when possible and to the extent legally allowed; for example, many facilities had begun providing information, counseling and consent sessions by phone rather than in person, and had switched to phone‐based follow‐up for medication abortion.

A few respondents indicated that these in‐clinic changes to reduce COVID‐19 risk were still not sufficient for people who were in high‐risk groups or had family members in high‐risk groups, and that they had taken additional steps to protect such individuals. These steps included having high‐risk individuals work from home or not work at all. A physician who was the former owner of a clinic in the West explained that the pandemic was an impetus for his own retirement: “Personally, I have retired in March due to age and COVID. I was already in process of retiring, but COVID moved that faster. My wife and I are high‐risk people.”

In addition to taking measures to reduce COVID‐19 risk, many respondents reported changes in the number of patients they were serving or the number of services they were providing. Although some mentioned a decline in patient volume resulting from the introduction of practices to ensure public health, such as spacing out patients to limit in‐person contact in the clinic, others reported increased volume, most commonly as a result of people traveling to their clinic from neighboring states where laws had limited the availability of abortion services. Some respondents mentioned changes in no‐show rates; for a few, rates fell, while for others they increased.

For some clinics, the pandemic had posed new logistical challenges, particularly within their supply chain. Disruptions had occurred not only for supplies of personal protective equipment, but also of routine commodities (e.g., paper towels, hand sanitizer, toilet paper and gloves) and medications (e.g., antibiotics). Clinics managed this issue in a variety of ways, including paying more for these supplies and seeking out alternatives, such as cloth masks. One respondent mentioned having resorted to bartering for supplies, and a few clinics had temporarily stopped providing services because of supply chain interruptions.

Some respondents mentioned that patients had been forced to navigate new logistical barriers related to travel. Some travel‐related difficulties were due to the general public health response to the pandemic, such as restrictions on travel across state lines; other patients had had trouble obtaining transportation because the support people who accompanied them could not come inside the clinic, and a respondent in a large city mentioned that patients were having difficulty getting to the clinic because of a lack of safe public transportation. A few respondents reported that patients were having difficulty arranging for childcare.

By far, the most common stressor mentioned by respondents that pertained to this theme concerned legal restrictions—specifically, whether states had designated abortion as a nonessential service during the early stages of the pandemic. Such designations, respondents noted, resulted in significant stress and confusion. For example, the manager of a clinic in the West wrote:

“Because of Governor [name's] EO [executive order] prohibiting elective surgeries, we have been on the edge of our seats to find out if the government will shut us down here. As of yet, we do not expect this—Governor [name] said that doctors may determine what counts as elective—but the anxiety is a torment.”

The effects of designating abortion as a nonessential health service were sometimes exacerbated by existing legal requirements. Some respondents, for example, explained that state laws requiring in‐person counseling or banning telemedicine for abortion impeded their ability to adopt some of the clinical and clinic‐flow innovations that they otherwise would have implemented. However, state laws were not always stressors. A few respondents noted that the laws in their state were buffers against what otherwise would have been an even more stressful situation. For example, one respondent mentioned that a recent state law that permitted nurse practitioners and nurse midwives to provide abortion care allowed them to continue providing services when the clinic's doctors were reassigned elsewhere.

A few respondents reported that harassment of their clinic staff and patients had continued or even increased since the beginning of the pandemic. A deputy director of a Midwest clinic wrote: “This has been hard, because protesters are still outside harassing patients. ‘Antis’ keep calling the Department of Health on us for COVID violations.” Conversely, other respondents mentioned receiving and appreciating community support. Some clinics had received thank‐you notes for their work, donations of masks, contributions to help pay for patients’ abortions, assistance with childcare from local medical schools, and financial support and clinical guidance from national organizations. However, a small number poignantly noted that abortion providers were not included in the well‐publicized thank‐you sentiments that people all over the country were delivering to other essential health workers.

Respondents mentioned three additional ways in which the early months of the pandemic may have lasting impacts on their clinics, their workforce and the services they provide. Themes related to these potentially long‐lasting effects were staff members’ having intense emotions without time to process them, financial impacts and (less commonly) opportunities arising from the pandemic.

The experience of intense emotions was a very common theme. These emotions, which respondents mentioned primarily when discussing the workforce, included fear and anxiety about getting sick with COVID‐19, and feeling unnerved about the apparent lack of a coordinated federal government response. They also reported that clinic personnel felt considerable stress and confusion related to changing clinical care and clinic flow protocols, to the new roles staff were taking on and to workforce changes (such as new hires and layoffs). Some respondents mentioned that they did not have sufficient time to process changes, and that they were just pushing through and feeling exhausted. For example, one administrator in the South wrote: “Stability is wobbly. Lack of time to process. Exhaustion.”

Respondents from clinics in states where the authorities had classified abortion as nonessential reported being further exhausted from the added stress of working in an environment where abortion access remained tenuous. The cumulative impact of navigating these two situations took a significant emotional toll on many, as the experience of a staff physician in a Southern state that declared abortion a nonessential service illustrates:

“We initially wanted to be proactive in using this as an opportunity to remove the medically unnecessary parts of the abortion appointments that were state legislated (ultrasound, state‐mandated information, waiting periods, in person medications etc.). [We were n]ot only unsuccessful in that venture but then were forced to shut down completely. The impact of shuffling patients and telling them that our doors may not be open the next day in the midst of back‐and‐forth legislation was a constant sense of chaos and helplessness.”

To help alleviate the intense emotions among staff, clinics took supportive actions, such as providing staff opportunities to talk about their emotions on a daily basis, providing daily lunches and even paying for therapy. Some respondents mentioned seeking to foster an emotional connection with staff. A few reported an appreciation for patients’ patience with longer wait times and patients’ expressions of gratitude that clinics had remained open. Others described negative emotional experiences, such as patients’ having to deal with legal and logistical uncertainties because of clinic closures related to state laws. An owner of a Midwestern clinic talked about the challenges of remaining “warm and fuzzy” with a patient while “sitting 6 feet away, and asking her to sanitize her hands before and after.” A few respondents mentioned that staff had less time with patients than previously, while others noted increased interactions with patients because of the new protocols.

Some comments concerned the pandemic's financial impact on clinics. Respondents mentioned a variety of increased costs, including those related to the hiring of new staff (or temporary staff to replace those in quarantine), hazard pay, adjusted leave policies, supplies (e.g., paper towels, toilet paper) and technology for staff working remotely. They also mentioned declines in revenue as a result of having to cancel or postpone appointments for services other than abortion and having to reduce the number of days they were providing abortions. The medical director of a clinic in the West wrote: “We have canceled all nonessential appointments or moved them to telephone calls and eventually video visits. This has led to extreme budgetary changes and the need to furlough clinician staff as well as other staff at every level of the organization.”

Finally, a very small number of respondents described the changes they had made to cope with the pandemic as opportunities and viewed some of these changes as worth keeping. An employee of a Northeastern clinic observed, “In some ways [the pandemic] has made it easier because we've been able to easily stop doing things that aren't medically necessary.” Another noted, “Because of the changes we've made, we have discovered how to be more efficient & will keep many of the changes adopted after the all clear from COVID‐19.” A third reported: “And most exciting, because of the limited availability of physicians who are needed for COVID‐specific care in their facilities, our nurse practitioner is now doing aspiration abortion on her own. Amazing!”

DISCUSSION

This rapid‐response study found that, regardless of region, independent abortion clinics’ workforces and financial sustainability have been affected by the COVID‐19 pandemic. Clinics have had to postpone or cancel health services, including abortion services, and have made substantial changes to their work schedules and clinic flow procedures.

While health care providers in general have been affected by the pandemic and resulting general public health responses, some abortion clinics have faced additional challenges. In particular, our study found that abortion clinics—and, by extension, abortion patients—in the South (and in some cases the Midwest) not only did not receive the government support that other health care facilities received during the pandemic, but were impacted by government responses that explicitly targeted abortion care. Respondents in these regions were more likely than those in other areas to report having had to temporarily close their clinics and to have canceled or postponed abortion services. This pattern of findings is consistent with results from a recent study that documented an increase in requests for medication abortion pills to an online service that supports people self‐managing abortions. In addition, it appears that some preexisting restrictive abortion policies—such as requiring in‐person visits and banning telemedicine for abortion—have gotten in the way of abortion clinics being able to provide services in a manner that reduces COVID‐19 risks.

Our findings suggest that abortion clinics have experienced short‐term financial needs as well as challenges in retaining and supporting their workforce. As our study was exploratory, we did not quantify the number of clinics with these needs or measure the intensity of need within clinics. Still, our findings indicate that abortion clinics are not notably different from other health care facilities in that these needs exist.25, 26, 27, 28 What is remarkable, though, is that despite facing significant challenges, the overwhelming majority of clinics in our sample—more than 80%—were able to continue providing abortion care during the early weeks of the pandemic in the United States. This resiliency and commitment to patients among the abortion providers we surveyed is consistent with findings from previous research with independent abortion providers —in particular, that to protect patients, providers often bear the brunt of restrictive abortion policies.

Going forward, a need remains for additional research to understand the impact that abortion service delays and clinic closures have on patients, particularly in the South and Midwest. Researchers should examine whether any clinics have closed permanently because of the COVID‐19 pandemic or the designations of abortion as a nonessential service. In addition, further research should examine whether the short‐term impacts on the abortion care workforce have persisted and assess current needs of these individuals. Finally, our findings suggest that in the future, public health teams that are planning pandemic responses might consider including abortion clinics in their efforts, as they do with other health clinics. For example, these planners could have systems in place to identify abortion clinics’ needs and capacities for continuing to provide abortion services, include clinics in updates regarding safety protocols to reduce health risks from the pandemic, and assess clinics’ financial needs to help ensure that the facilities are sustainable over time.

Limitations and Strengths

There are a number of limitations to note. First, it is not obvious how to characterize the response rate to the survey, as we lack information about the number of clinics we may have reached with our email messages about the study. We estimate, though, that our recruitment emails were sent to about 150–165 unique independent clinics; thus, about 60% of clinics we reached out to completed the survey. As of November 2019, there were fewer than 350 independent abortion clinics in the United States, so our respondents, who provided information on 103 clinics from all regions of the country, likely represent a significant proportion of independent abortion clinics, although our findings may not be generalizable to clinics not affiliated with either the ACN or the Abortion Clinical Research Network. While we have likely achieved our aim of documenting the range of ways that the pandemic, the general public health response and the designation of abortion as a nonessential service have affected abortion clinics, it is possible that the impacts were different at clinics that did not participate or that we did not reach with our outreach efforts. In particular, our sample may not include clinics that closed for longer periods—or permanently—as a result of the pandemic. In addition, contrary to our original intent, it appears that some of the respondents represented abortion clinics that were part of academic medical centers, and thus may not be representative of freestanding independent clinics. Similarly, we excluded Planned Parenthood clinics from our study. While such clinics provide about one‐third of abortions in the United States, their experiences may differ from those of independent clinics, as additional stressors in the form of government policy sometimes single out Planned Parenthood clinics.32, 33 However, Planned Parenthood clinics may also have buffers, such as the support of a large organization, that could allow them to centralize some activities, including updating clinical protocols and clinic‐flow protocols and training staff about new protocols.

Second, because our survey did not include an item asking respondents to identify the nonabortion services their clinics had provided in 2019, we were unable to assess whether a clinic that did not report having to postpone or cancel such services owing to COVID‐19 had actually provided them prior to the pandemic. We do note, however, that the findings for abortion services did not change when we conducted a sensitivity analysis that was restricted to clinics that indicated they had provided those services. Third, we included only one response per clinic. While we requested that a clinic manager or director complete the survey on behalf of their clinic, we did not restrict participation to persons in these roles. As a result, we had responses from individuals working in a range of roles in the clinics, and respondents’ experiences and their knowledge of the pandemic's impact may have differed according to their role. Fourth, to help ensure privacy, we asked respondents to report the region in which their clinic was located, but not the state; thus we were unable to determine which clinics were in locations that had experienced the early surges of the pandemic and which were in locations that had had few cases at the time of the survey. Therefore, we cannot distinguish the chaos resulting from the early surges in some specific locations from the more generalized efforts to navigate the early weeks of the pandemic. Finally, we collected responses over a five‐week period, and clinics that participated later in the recruitment window may have had more time to have one of the experiences we asked about (e.g., clinicians or staff getting sick, having to postpone or cancel patient appointments). Thus, these findings should be viewed as the floor, rather than the ceiling, in terms of the proportion of clinics that had these experiences.

Our study also has a number of strengths. Notably, this was a rapid‐response study that captured the real‐time experiences of independent abortion clinics during the height of chaos and challenges in the initial weeks of the COVID‐19 pandemic in the United States. Moreover, despite the brevity of the study period, we were able to collect data on more than 100 clinics, which together represented a wide range of locations, sizes and services provided.

Conclusions

Independent abortion clinics have been deeply affected by the COVID‐19 pandemic; they have had to navigate the challenges posed by the pandemic itself, the unintended consequences of general public health responses and the direct consequences of designations of abortion as a nonessential service. They have experienced significant disruptions to their ability to provide care, and these disruptions have been especially common in the South and Midwest, where states were particularly likely to designate abortion as a nonessential service. Yet despite these challenges, most clinics continued to provide abortion care. Additional support may be needed to build on this short‐term resilience to ensure the sustainability of independent abortion clinics and the well‐being of their workforce.