COVID-19 impacts on abortion care-seeking experiences in the Washington, DC, Maryland, and Virginia regions of the United States.

CONTEXT: Many people wanted to avoid or delay childbearing during the COVID-19 pandemic. This study sought to examine the extent COVID-19 influenced abortion care-seeking in a region that did not enact policy restricting abortion due to the pandemic, has high service availability, and few abortion-restrictive policies.
METHODOLOGY: We conducted telephone surveys with adults (n=72) requesting abortion appointments between September 2020 and March 2021 at five clinics in Washington, DC, Maryland, and Virginia. We used χ2 tests to compare sociodemographic, reproductive history, service delivery characteristics, and pandemic-related life changes by whether COVID-19 influenced abortion care-seeking.
RESULTS: Most respondents (93%) had an abortion at the time of the survey, 4% were awaiting their scheduled appointment, and 3% did not have an appointment scheduled. Nearly 40% of people reported COVID-19 influenced their decision to have an abortion. These individuals were significantly more likely to report "not financially prepared" (44% vs. 16%) as a reason for termination compared to people reporting no influence of COVID-19. They were also more likely to have lost or changed their health insurance due to pandemic-related employment changes (15% vs. 2%), report substantial money difficulties due to COVID-19 (59% vs. 33%), and report that paying for their abortion was "very difficult" (25% vs. 2%).
CONCLUSION: COVID-19 influenced many people to have an abortion, particularly those financially disadvantaged by the pandemic. Expansion of Medicaid abortion coverage in Washington, DC and Virginia could reduce financial barriers to care and help people to better meet their reproductive needs amid future crises.

INTRODUCTION

Since the World Health Organization declared COVID‐19 a pandemic in March 2020, the continued spread of the coronavirus has magnified health system, economic, and social disruption worldwide. , In addition to impacts on morbidity and mortality in the United States (US), , the pandemic likely exacerbated existing inequities in contraception and abortion access, , further constraining reproductive autonomy by limiting people's power to decide and control contraceptive use, pregnancy, and childbearing. Research has shown critical impacts of COVID‐19 on sexual and reproductive healthcare, including service disruptions, increased healthcare expenses, and decreased patient ability to pay for care. , , At the same time, pandemic‐induced financial uncertainty influenced desires to delay or avoid pregnancies, underscoring the need to strengthen access to contraception and abortion care. Moreover, not obtaining a wanted abortion is associated with adverse outcomes after delivery, including increased household poverty, increased food and housing instability, and an increased likelihood of staying in an abusive relationship compared to people who can obtain an abortion. , , Yet long before COVID‐19, many people faced significant barriers to care due to difficulties coordinating travel, raising funds, and locating a provider. , , Disproportionate pandemic‐related financial impacts on people experiencing existing systemic inequities, including those with low incomes and racial/ethnic minorities who have experienced the most substantial job losses, likely compound these existing barriers. ,

At the height of COVID‐19 restrictions, some states instituted policies to restrict procedures deemed “elective and non‐essential,” effectively impeding time‐sensitive abortion care and focusing legislative attention and social science research on states considered most hostile to abortion access. , , , , , Consequently, scant evidence exists regarding abortion care‐seeking during heightened pandemic restrictions, economic instability, and health service disruptions in regions comparatively more supportive of abortion access. COVID‐19 may have nevertheless posed significant barriers to abortion care‐seeking in supportive policy contexts. This study aimed to examine the extent COVID‐19 influenced people's abortion decisions and investigate the factors associated with terminating a pregnancy due to COVID‐19 by surveying patients seeking abortion care in Washington, DC, Maryland, and Virginia, a setting less constrained by access to abortion services. ,

METHODS

Study setting

Legislators in Washington, DC, Maryland, and Virginia did not use COVID‐19 as an opportunity to enact further abortion policy restrictions like many other states. Instead, Washington, DC, Maryland, and Virginia had existing abortion‐supportive laws and introduced others during the pandemic. , Maryland had the most supportive abortion laws among the three study settings, including Medicaid coverage and 44 facilities providing abortion services. Both Washington, DC and Virginia had limited Medicaid coverage for abortion, but eight and 32 facilities providing abortion care, respectively.

Recruitment

We recruited patients from five abortion clinics in Washington, DC, Maryland, and Virginia—one Washington, DC clinic, two Maryland clinics, and two Virginia clinics. Two of the recruitment clinics were hospital‐based, and three were free‐standing clinics. The clinics that agreed to participate in the study came from a pool of 11 regional clinics that, at the outset of COVID‐19, were participating, along with two study investigators, in weekly regional abortion service strategy calls.

All English‐speaking adults ages 18 and older who contacted study clinics for abortion care between September 2020 and March 2021 to schedule an abortion appointment were eligible to participate. People were eligible to participate regardless of whether they had received an abortion when they completed the survey. Clinic staff provided study information during appointment scheduling calls or at the appointment time. Interested patients agreed to be contacted by the study team for a 15‐minute telephone survey (Appendix A).

Trained women research assistants called people who expressed initial interest within 2 weeks after they contacted a study clinic to schedule an abortion appointment. Respondents provided oral informed consent and received USD15 for their participation. We used REDCap to collect and manage survey data. , The Johns Hopkins Bloomberg School of Public Health Institutional Review Board at Johns Hopkins University approved study procedures.

COVID‐19 factors

We measured COVID‐19's influence on abortion care‐seeking using the question, “Did the COVID‐19 pandemic influence your decision to have an abortion?” Respondents answered “yes,” “no,” or “I don't know.” Due to the small number (n = 1) reporting “I don't know,” we combined this response with “no,” resulting in a dichotomous variable. We measured COVID‐19‐related financial strain on a 5‐point scale (1 = no money difficulties to 5 = significant difficulties); we grouped responses in analysis (1 = none; 2–3 = some; 4–5 = significant). Additional variables included binary measures of any loss/change in insurance coverage due to COVID‐19 relatedemployment changes and any diagnosed/suspected coronavirus infection.

Sociodemographic, reproductive, and service‐delivery characteristics

Sociodemographic variables included age, gender identity, race/ethnicity, sexual orientation, education, change in respondent's and partner's employment since February 2020, relationship status, residence, number of children, and prior abortion. We also asked respondents whether they had considered self‐inducing and abortion or had attempted to self‐induce and asked, “What are the reasons that you decided to have an abortion?” Interviewers selected from a list of response options informed by previous research , or selected “Other.”

Surveys also captured several characteristics about the clinic encounter, including clinic location, number of clinics respondents called, ability to schedule an appointment at the study clinic, and gestational age at initial clinic contact and appointment. Additionally, people who had an abortion reported abortion type, time between initial clinic contact and appointment, payment method, level of payment difficulty, and satisfaction with care.

Analysis

We conducted descriptive analyses to examine sample characteristics and reported COVID‐19 related impacts on abortion decision‐making. We used χ 2 tests to examine differences between those who did and those who did not report COVID‐19 influenced their decision to terminate by sociodemographic, reproductive, and abortion service characteristics. In supplementary analyses, we examined select abortion and COVID‐19 characteristics by race. We conducted all analyses in Stata version 16.1.

RESULTS

Sample

Clinic staff approached a total of 412 people and 35% (n = 144) indicated an initial interest in participation; we were unable to contact 31% (n = 42) of this group. Among the 102 people the study team could contact, 71% (n = 72) agreed to participate (Table 1). The final sample included 13% (n = 9) recruited from the Washington, DC study clinic, 35% (n = 25) from Maryland clinics, and 53% (n = 38) from Virginia clinics (Table 2). Over half of respondents (n = 37) resided in Virginia, 40% (n = 28) lived in Maryland, and none lived in Washington, DC. Most respondents (n = 67, 93%) had an abortion at the time of the survey, 4% (n = 3) were awaiting their scheduled appointment, and 3% (n = 2) did not have an appointment scheduled (Table 2).

TABLE 1

Sociodemographic and reproductive characteristics of respondents seeking an abortion in the Washington, DC, Maryland, and Virginia region overall and by whether COVID‐19 influenced their decision to have an abortion (N = 72)

Characteristic	Overall N = 72; n (col %)	COVID‐19 influenced respondent's decision to have an abortion n (%)		p value a
		Yes (n = 27)	No (n = 45)
Age (in years)				0.84
18–20	10 (14)	3 (11)	7 (16)
20–29	32 (44)	11 (41)	21 (47)
30–39	28 (39)	12 (44)	16 (36)
≥40	2 (3)	1 (4)	1 (2)
Gender identity
Woman	72 (100)	27 (100)	45 (100)
Race/ethnicity				0.82
White, non‐Hispanic	29 (40)	10 (37)	19 (42)
Black, non‐Hispanic	28 (39)	11 (41)	17 (38)
Asian/Pacific Islander	5 (7)	3 (11)	2 (4)
Hispanic/Latinx	4 (6)	1 (4)	3 (7)
Other/multiracial	6 (8)	2 (7)	4 (9)
Sexual orientation				0.18
Heterosexual or straight	65 (90)	26 (96)	39 (87)
Homosexual, gay, or lesbian	0 (0)	–	–
Bisexual	7 (10)	1 (4)	6 (13)
Highest level of education				0.15
Less than a 4‐year degree	50 (69)	16 (59)	34 (76)
4‐year degree and above	22 (31)	11 (41)	11 (24)
Employment Status				0.46
Unemployed	28 (29)	13 (48)	15 (33)
Part‐time	10 (14)	3 (11)	7 (16)
Full‐time	34 (47)	11 (41)	23 (51)
Currently attending school	25 (35)	12 (44)	13 (29)	0.18
Change in employment since February 2020				0.50
No change	45 (63)	16 (59)	29 (64)
Lost job or reduced hours	19 (26)	9 (33)	10 (22)
Increase in employment hours	8 (11)	2 (7)	6 (13)
Relationship status at initial clinic contact				0.56
Single	28 (39)	10 (37)	18 (40)
In a relationship, not cohabitating	10 (14)	2 (7)	8 (18)
Cohabitating	20 (28)	10 (37)	10 (22)
Married	12 (17)	4 (15)	8 (18)
Other/prefer not to say	2 (3)	1 (4)	1 (2)
Partner's employment status b				0.56
Unemployed	8 (19)	3 (19)	5 (19)
Part‐time	5 (12)	3 (19)	2 (8)
Full‐time	29 (69)	10 (63)	19 (73)
Partner's change in employment since February 2020 c				0.96
No change	33 (83)	13 (81)	20 (83)
Lost job or reduced hours	5 (13)	2 (13)	3 (13)
Increase in employment hours	2 (5.0)	1 (6)	1 (4)
Number of children				0.76
None	36 (50)	12 (44)	24 (53)
1–2 children	26 (36)	11 (41)	15 (33)
3–5 children	10 (14)	4 (15)	6 (13)
Had a previous abortion prior to March 2020	21 (29)	10 (37)	11 (24)	0.26
Residence				0.34
Washington, DC	0 (0)	0 (0)	0 (0)
Maryland	28 (39)	10 (37)	18 (40)
Virginia	37 (51)	16 (59)	21 (47)
Other d	7 (10)	1 (4)	6 (13)

Based on Chi‐squared tests.

Based on data from n = 42 respondents who reported being in a relationship at initial clinic contact.

Based on N = 40 respondents who reported on the characteristics.

Other residence includes West Virginia, North Carolina, Florida, Pennsylvania, Nebraska, and Ontario.

TABLE 2

Abortion trajectory and abortion characteristics of respondents seeking an abortion during COVID‐19 in the Washington, DC, Maryland, and Virginia region overall and by whether COVID‐19 influenced their decision to have an abortion (N = 72)

	Overall N (col %)	COVID‐19 influenced the respondent's decision to have an abortion n (%) or mean (SE)
		Yes (n = 27)	No (n = 45)	p value a
Had abortion				0.52
Yes	67 (93)	24 (89)	43 (96)
Not yet, but appointment scheduled	3 (4)	2 (7)	1 (2)
No appointment scheduled at time of survey	2 (3)	1 (4)	1 (2)
Type of abortion b				0.72
Medication	26 (39)	10 (42)	16 (37)
Aspiration/surgical	41 (61)	14 (58)	27 (63)
Received preferred type of abortion b				0.21
Yes/no preference	60 (90)	20 (83)	40 (93)
No	7 (11)	4 (17)	3 (7)
Satisfaction with care received (5 = fully satisfied) b	4.90 (0.24)	4.95 (0.21)	4.92 (0.28)	0.91
Abortion payment b
Out‐of‐pocket with own money	35 (52)	10 (42)	25 (58)	0.20
Out‐of‐pocket with money from family and/or friends	6 (9)	2 (8)	4 (9)	0.89
Private insurance	14 (21)	5 (21)	9 (21)	0.99
Public insurance	13 (19)	4 (17)	9 (21)	0.67
Assistance from an abortion fund	20 (30)	8 (33)	12 (28)	0.64
Difficulty paying for abortion b				0.03
Not at all difficult	24 (36)	6 (25)	18 (42)
Not very difficult	17 (25)	6 (25)	11 (26)
Somewhat difficult	19 (28)	6 (25)	13 (30)
Very difficult	7 (11)	6 (25)	1 (2)
Recruitment clinic location				0.21
Washington, DC	9 (13)	1 (4)	8 (18)
Maryland	25 (35)	10 (37)	15 (33)
Virginia	38 (53)	16 (59)	22 (49)
Reasons for seeking abortion (select multiple)
Not financially prepared	19 (26)	12 (44)	7 (16)	0.01
Not the right time for a baby	38 (53)	16 (59)	22 (49)	0.39
Partner‐related reasons	13 (18)	6 (22)	7 (16)	0.48
Need to focus on other children	15 (21)	8 (30)	7 (16)	0.16
Interferes with future opportunities (e.g., work, school)	14 (19)	8 (30)	6 (13)	0.09
Not emotionally or mentally prepared	11 (15)	5 (19)	6 (13)	0.55
Health‐related reasons (pregnant person or fetus)	20 (28)	5 (19)	15 (33)	0.17
Wanted a better life for the baby than could provide	2 (3)	1 (4)	1 (2)	0.71
Considered self‐inducing an abortion	5 (7)	3 (11)	2 (4)	0.28
Attempt to self‐induce an abortion	1 (1)	1 (100)	0 (0)	–

Based on Chi‐squared tests for categorical variables and Wilcoxon Rank Sum Test for satisfaction with care.

Based on respondents who received an abortion at time of survey (n = 67).

Approximately 10% (n = 7) of respondents who had an abortion came from outside the region, including other US states and Canada. None of these characteristics differed significantly by whether COVID‐19 influenced the respondent's decision to have an abortion (Table 1). Of people who paid out‐of‐pocket for their abortion, the majority were white (p < 0.01; Table B1) and 85% of respondents who used public insurance to pay were Black (p < 0.01; Table B1). Forty‐three percent of people who reported paying for their abortion was “very difficult” identified as Black compared to 29% white, although results did not differ significantly by race (p = 0.19; Table B1).

TABLE B1

Select characteristics of respondents who sought abortion care during COVID‐19 by race, Washington, DC, Maryland, and Virginia (N = 72)

Characteristic n (row %)	White n (row%)	Black n (row%)	Multiracial/other n (row%)	p value a
Recruitment clinic location				<0.01
Washington, DC	5 (56)	2 (22)	2 (22)
Maryland	3 (12)	19 (76)	3 (12)
Virginia	21 (55)	7 (18)	10 (26)
Had abortion				0.55
Yes	28 (42)	25 (37)	14 (21)
Not yet, but appointment scheduled	1 (33)	2 (66)	0 (0)
No appointment scheduled at time of survey	2 (67)	1 (33)	0 (0)
Type of abortion b				0.54
Medication	13 (50)	8 (31)	5 (19)
Surgical	15 (37)	17 (41)	9 (22)
Received preferred type of abortion b				0.29
Yes/no preference	27 (45)	21 (35)	12 (20)
No	1 (14)	4 (57)	2 (29)
Abortion payment b
Out‐of‐pocket with own money	20 (57)	7 (20)	8 (23)	<0.01
Out‐of‐pocket with money from family and/or friends	3 (50)	1 (17)	2 (33)	0.48
Private insurance	6 (43)	5 (36)	3 (21)	0.96
Public insurance	1 (8)	11 (85)	1 (8)	<0.01
Assistance from an abortion fund	8 (40)	7 (35)	5 (25)	0.84
Difficulty paying for abortion b				0.19
Not at all difficult	7 (29)	14 (58)	3 (13)
Not very difficult	9 (53)	4 (24)	4 (24)
Somewhat difficult	10 (53)	4 (21)	5 (26)
Very difficult	2 (29)	3 (43)	2 (29)
Diagnosed or suspected COVID‐19 infection at any point
Yes	11 (48)	8 (35)	4 (17)	0.67
No	18 (37)	20 (41)	11 (22)
Lost insurance coverage/changed plans due to COVID‐19‐related changes in employment				0.49
Yes	2 (40)	1 (20)	2 (40)
No	27 (40)	27 (40)	13 (19)
COVID‐19 caused money difficulties for self and immediate family				0.44
No difficulties	6 (50)	3 (25)	3 (25)
Some difficulties	14 (48)	11 (38)	4 (14)
Significant difficulties	9 (29)	14 (45)	8 (26)

Based on Chi‐squared tests.

Based on respondents who received an abortion at time of survey (n = 67).

COVID‐19 impacts

Overall, 38% (n = 27) of people reported the COVID‐19 pandemic influenced their decision to have an abortion (Table 1). In addition, one in four respondents lost their job or were forced to reduce hours compared to pre‐COVID levels in February 2020; 13% (n = 5) reported their partner lost their job or reduced hours during the same time. Nearly 40% (n = 26) reported payment was somewhat or very difficult. This was significantly more common among those who said COVID‐19 influenced their decision to have an abortion than among people whose decisions were not affected by COVID‐19 (50% vs. 32%; p = 0.03). While approximately one in four respondents indicated financial unpreparedness as a reason for terminating, this reason was significantly more common among those who reported COVID‐19 influenced their decision (44% versus 16%, p = 0.1); Similarly, 19% (n = 14) indicated the pregnancy would interfere with future opportunities. This was not significantly more common among those for whom COVID‐19 was a factor in their termination decision than those for whom it was not (30% versus 13%, p = 0.09). No respondents indicated “Other” reasons for having an abortion not listed on the survey.

Overall, 7% (n = 5) lost or changed insurance coverage, but those who reported COVID‐19 influenced their decision to terminate were significantly more likely to have experienced insurance changes than those who did not report COVID‐19 influenced their decision (Table 3; 15% versus 2%, p = 0.04). In addition, while 43% (n = 31) of respondents reported significant COVID‐19‐related money difficulties, those who said COVID‐19 influenced their decision to seek abortion were significantly more likely to experience financial issues (59% vs. 33%, p = 0.03). Thirty‐two percent of respondents (n = 23) reported diagnosed or suspected COVID‐19 at any point during the pandemic, which was not different by whether COVID‐19 influenced the decision to terminate.

TABLE 3

COVID‐19‐related impacts on people seeking abortion in the Washington, DC, Maryland, and Virginia region overall and by whether the pandemic influenced respondent's decision to have an abortion (N = 72)

		COVID‐19 influenced the decision to have an abortion
	Overall N (col %)	Yes (N = 27)	No (N = 45)	p value a
Lost insurance coverage/changed plans due to COVID‐19‐related changes in employment				0.04
Yes	5 (7)	4 (15)	1 (2)
No	67 (93)	23 (85)	44 (98)
COVID‐19 caused money difficulties for self and immediate family				0.03
No difficulties	12 (17)	1 (4)	11 (24)
Some difficulties	29 (40)	10 (37)	19 (42)
Significant difficulties	31 (43)	16 (59)	15 (33)
Diagnosed or suspected COVID‐19 infection at any point				0.47
Yes	23 (32)	10 (37)	13 (29)
No	49 (68)	17 (63)	32 (71)

Based on Chi‐squared tests.

DISCUSSION

The reasons people sought abortion care mainly were consistent with reasons described in pre‐pandemic US studies , ; however, nearly two in five respondents indicated COVID‐19 influenced their decision to have an abortion. Over half of respondents paid‐out‐of‐pocket for their abortion; however, those who COVID‐19 influenced to have an abortion experienced the most difficulty paying for care, more frequently sought abortion due to financial constraints, and reported more significant loss or change in insurance coverage compared to those for whom COVID‐19 was not a factor in their decision. These findings suggest people who said that COVID‐19 influenced their decision to have an abortion were more disadvantaged by pandemic economic constraints and experienced the most money difficulties. Thus, COVID‐19 likely intensified financial limitations preventing people from feeling they could continue a pregnancy. The ability to decide whether and when to give birth is inextricably linked to economic stability and empowerment. , Thus, lack of accessible abortion care, particularly during times of large‐scale economic disruption, threatens people's lives and futures. , ,

Our findings suggest economic challenges may have influenced people's decision to have an abortion when faced with an unintended pregnancy during COVID‐19 and align with previous research. Studies show disproportionate pandemic disadvantages contributed to changes in fertility desires and access to reproductive healthcare, especially among groups already experiencing systematic economic and racial inequities. , , People financially disadvantaged in early 2020 reported wanting to avoid or delay childbearing because of COVID‐19. , In one study, 40% of respondents changed their plans about the timing and number of children desired due to COVID‐19, with the greatest change in preferences seen among Black, Hispanic, and low‐income respondents. Another study suggests Black and Hispanic women experienced greater food, transportation, and housing insecurity during COVID‐19 compared to white women, which was associated with an increased desire to avoid pregnancy. At the same time, pandemic job and income losses diminished people's ability to meet their contraceptive needs, likely influencing abortion care‐seeking. Expansion of Medicaid support for abortion in Washington, DC and Virginia could reduce financial barriers to care and help people to better meet their reproductive needs amid future crises. Finally, continuing research on the impacts of COVID‐19 on abortion care‐seeking in both restrictive and supportive policy environments is necessary for informed care and policymaking.

Readers should interpret these findings with caution. Results may not represent all abortion patients at Washington, DC, Maryland, and Virginia clinics or outside the region; those experiencing the most substantial COVID‐19 financial and health impacts may have been less likely to have the time to participate. Alternatively, they may have been more motivated to participate given the small financial incentive. Thus, the direction of the bias our selective sample includes is difficult to determine. Additionally, only a small number of people unable to schedule an appointment are included in this study, potentially underestimating the association and impact of COVID‐19 on abortion care‐seeking experience. It is possible that people who were unable to access a wanted abortion and those unable to contact an abortion clinic experienced insurmountable barriers to accessing care. Finally, the small sample size limits our power to detect statistically significant differences in sample sociodemographic characteristics we might anticipate given previous research , , on pandemic exacerbated racial and socioeconomic inequities.

This study nevertheless contributes to emerging evidence on the impacts of COVID‐19 on sexual and reproductive health. We provide timely data from an area with high abortion service availability and few policies restricting abortion access. Findings support an association between financial instability and having an abortion due to pandemic‐induced conditions, which researchers should explore further in studies of abortion access, demand, and care‐seeking. Qualitative work could help elucidate the myriad factors and potential confounders that motivate people to seek abortion due to pandemic‐related financial stresses not captured in this study, such as childcare challenges, increasing debt, difficulty affording food, and housing insecurity. ,

In conclusion, while faced with heightened existing barriers to abortion care, the pandemic economic environment likely impacted reproductive trajectories, specifically abortion care‐seeking among financially disadvantaged people.

FUNDING INFORMATION

This research was supported by a Johns Hopkins University Alliance for Healthier World's 2020 COVID‐19 Launchpad grant. The funder had no role in the study design, data collection, analysis, writing of the report, or deciding to submit the article for publication.