“In some clinics, they said it’s elective, and then they would refuse”: A Mixed-Methods Study on the impact of the COVID-19 pandemic on access to abortion services in Germany

Objective The World Health Organization recognizes abortion as essential health care and has encouraged governments to ensure access to abortion services throughout the COVID-19 pandemic. However, the threat of infection combined with government responses to COVID-19 have impacted access to abortion services globally. This study explores access to abortion in Germany during the pandemic. Methods This study used a mixed-methods design. An analysis of data collected by Women on Web (WoW) was carried out to assess women’s reasons for choosing telemedicine abortion outside the formal health system in Germany during the pandemic. Descriptive statistics were generated for 2057 requests for telemedicine abortion received by WoW between March 2020-March 2021. Semi-structured interviews were conducted with eight healthcare professionals involved in the provision of abortion services to explore how they perceive of women’s access to abortion services in Germany during the pandemic. Results The quantitative analysis found that preferences and needs for privacy (47.3%), secrecy (44.4%) and comfort (43.9%) were the most common reasons for choosing telemedicine abortion. COVID-19 was another important reason (38.8%). The thematic analysis of the interviews was organized into two overarching themes: service provision, and axes of difference. Conclusions The pandemic affected the provision of abortion services as well as the circumstances of women seeking abortion. The main barriers to access were financial constraints, privacy issues, and lack of abortion providers. Throughout the pandemic, accessing abortion services was more difficult for many women in Germany, especially women experiencing multiple and overlapping forms of discrimination.


Introduction
The COVID-19 pandemic and the actions taken by governments to prevent the spread of the virus have significantly impacted access to sexual and reproductive health services globally [1]. As past global health emergencies have shown, sexual and reproductive health and rights (SRHR), particularly access to safe abortion, are often neglected during crises [1,2]. The pandemic has placed a spotlight on the existing barriers to abortion access worldwide, particularly those experienced by women who are structurally disadvantaged because of socioeconomic status, chronic illness, caring responsibilities, or other social factors [2,3]. The pandemic has also created new barriers to accessing abortion services [3].
Women on Web (WoW) is a non-profit organization that provides safe, accessible, and affordable telemedicine abortion to women around the world. Telemedicine abortion refers to the use of information and communication technologies to self-manage medical abortion at home [4]. WoW opened its services in Germany in April 2019 [5]. The number of requests for telemedicine abortion rapidly increased in the first year, suggesting that there is an unmet need for abortion services in Germany [5]. Despite the relatively liberal abortion law, inequities and barriers hamper access to abortion services [6,7].
A few previous studies have investigated different aspects of access to abortion services in Germany during the pandemic, including abortion policy responses, but none of the studies provide both the perspectives of providers and abortion seekers [8][9][10]. The aim of this study is therefore to provide a more comprehensive understanding of how the pandemic has impacted access to abortion services in Germany. There are two specific objectives: 1. To assess women's 1 reasons for choosing telemedicine abortion outside the formal health system in Germany during the COVID-19 pandemic. 2. To explore how healthcare professionals involved in the provision of abortion services 2 in Germany perceive of women's access to abortion services during the COVID-19 pandemic.

Theory
This study applies the terms structural violence and intersectionality to make sense of the data in a broader theoretical context. Structural violence, introduced by Johan Galtung in 1969 [11], represents social injustice inflicted by authorities and state governance through social, economic and health-related policies [12] and the failure to mitigate severe consequences of such policies among the poorer and vulnerable part of the population. Intersectionality is a theoretical tool to understand the how social categories such as ethnicity, gender, class, sexual orientation, age, and disability, interact in assembling disadvantage or privilege for the individual [13]. Instead of assuming that women as a homogenous group experience access to abortion the same way, we assess what circumstances impact access to abortion, and reasons for choosing telemedicine, during the pandemic.

Study design
A mixed-methods study design was used to gain an in-depth understanding of the impact of the pandemic on access to abortion services in Germany. The quantitative data was obtained from WoW in the beginning of April 2021. To complement the quantitative data, the perspectives of German healthcare professionals (HPs) were gathered through semi-structured interviews between April and June 2021. We used methodological triangulation to increase reliability. The quantitative and qualitative results were compared to nuance findings and to identify contradictions.

Study setting
In Germany, most abortions are illegal under Section 218 and 219 of the German Criminal Code but exempt from punishment under certain circumstances [14,15]. Abortion within the first twelve weeks after conception is unlawful but unpunishable if the woman attends counselling at a government authorized counselling center at least three days prior to the abortion 3 . An abortion costs minimum 360 euros, but women with low or no income can apply for the federal state to cover the costs. The application process consists of several steps, and the application must be filed prior to the abortion [15]. Self-management of abortion is not punishable [14].
In 2020, 99,948 abortions were performed in Germany, of which 96,110 abortions included counselling services [16]. In 2019, Germany had 5.8 abortions per 1,000 women aged 15 to 49 years, compared to an average of 17 in Europe and Northern America [17,18]. In response to the pandemic, the German Federal Government implemented three nationwide lockdowns between March-May 2020, November 2020-January 2021, and January-March 2021. Some of the main COVID-19 restrictions and measures implemented included social distancing measures, postponing non-urgent procedures, travel bans, border closures, movement restrictions, and closure of schools, kindergartens, and non-essential stores [19][20][21][22][23][24]. The Government did allow counselling centers to provide mandatory pre-abortion counselling via telemedicine to compensate for the lockdown.

Survey data collection and analysis
Women can request a telemedicine abortion by completing an online consultation form 4 , which includes demographic questions about age and place of residence, medical questions about gestational age, contraceptive behavior, and medical contraindications, and research questions about reasons for wanting an abortion and for choosing telemedicine abortion through WoW. When submitting the online consultation form, women agree with the Terms of Use 5 and thereby consent to the fully anonymized use of their information for scientific research purposes. Consent can be withdrawn at any time by emailing WoW. Hence fully anonymized data was obtained from WoW. The dataset included a total of 2078 requests for telemedicine contraception and abortion between 18 March 2020 and 18 March 2021. All requests for telemedicine abortion were analyzed regardless of the approval status. Requests for intrauterine devices and emergency contraceptive pills and requests from women not residing in Germany were excluded; final inclusion n = 2057. Descriptive statistics were generated in Microsoft Excel to describe the basic features of the data in the online consultations.

Qualitative interview data collection and analysis
Semi-structured, open-ended interviews were conducted online with eight HPs: three abortion providers and three pre-abortion counsellors in Germany (Berlin n = 5, Hesse n = 1), as well as a member of WoW's helpdesk and a member of WoW's medical team ( Table 1).
The HPs were mainly found through the information service on family planning provided by the Federal Centre for Health Education 6 and the official website of the state of Berlin 7 . Approximately 150 HPs across Germany were contacted, of which six agreed to participate. The counsellors had to work at a state approved counselling center. The WoW helpdesk member answers emails from abortion-seekers in German, while the WoW medical team member reviews consultations in Table 1 Background information on the interview participants. General practitioner WoW (medical team) n/a 1 "Women" is used while acknowledging that transmen, nonbinary persons, and other individuals may also want or need an abortion (3). 2 In this study, the term "abortion services" refers to all health services associated with termination of pregnancy, including pre-abortion counselling and medical and surgical abortion procedures. 3 The law and policies concerning abortions when the pregnancy is caused by a sexual offence or if the pregnancy poses a risk to the health of the woman differ. 4 https://www.womenonweb.org/en/i-need-an-abortion. 5 https://www.womenonweb.org/en/page/493/terms-of-use-and-privacy-p olicy. 6 https://www.familienplanung.de. 7 https://www.berlin.de.
German. The interviews were conducted in English, lasted 20-50 min, and were audio recorded. Written informed consent was obtained from all participants. The interview guide covered topics including personal experiences in providing abortion services before and during the pandemic, and perceptions of how the pandemic has affected abortion providers, patients, and access to abortion services in Germany.
Inductive thematic analysis was conducted at the semantic level with a realist approach, following the six phases of thematic analysis outlined by Braun and Clarke [25]. The first researcher (HR) reported the experiences and meanings of the participants and did not try to fit the data within a pre-existing theoretical framework, but associated the data with central concepts [25]. The analysis was discussed with a second researcher (BT). The first researcher (HR) is a white, cisgender, ablebodied, middle-class woman from Norway with first-hand experience of medical abortion. The researcher continuously engaged in reflexivity to account for her situatedness within the research and the effect it may have on the knowledge produced.

Quantitative
The mean age of the respondents was 28.94 years old. The youngest and oldest respondents were 14 and 51 years old, respectively. WoW received the highest number of requests in April 2020 (n = 249) and January 2021 (n = 204). Background characteristics of participants and categorical reasons for requesting telemedicine abortion are shown in Tables 2 and 3.

Qualitative
Two overall themes were identified: (1) service provision, and (2) axes of difference.
Service provision All the HPs witnessed that accessing abortion services was more difficult for women during the pandemic. Accessing mandatory preabortion counselling (hereafter counselling) was more difficult at the beginning of the first lockdown, when the counselling centers could not offer face-to-face counselling and had not begun offering telephone and video counselling yet. While some counselling centers quickly began offering telephone and video counselling, others never did. Several of the HPs mentioned that timely access to counselling was an issue during the pandemic.
So, it might have taken a little longer to get the counselling […] more effort, and calling more centers, but it was not impossible. (HP 4) […] and perhaps it needed a bit more time arranging everything. So sometimes it can be quite fast, you do the counselling on Monday, you have to wait three days -Tuesday, Wednesday, Thursdayon Friday you can do the medical treatment, or you could perhaps have the surgery, and I think this fastness was not given at any time last year. (HP 5) The HPs mentioned several reasons why it was difficult to access abortion services during the pandemic. Fewer clinics and hospitals offered abortion services due to lack of childcare, risk of infection, and abortion views. As the schools and kindergartens were closed, many abortion providers stopped working or worked less due to lack of childcare. At the onset of the pandemic, many senior abortion providers stopped working because they were at higher risk of serious illness. Additionally, many abortion providers stopped performing abortions because they categorize abortion as a non-essential health service: in some clinics, they said it's elective, and then they would refuse (HP 7). * Some of the questions are optional, which explains the missing data. ⊥ Multiple responses allowed: total response percentages therefore exceed 100%. In the first 3-4 months of the pandemic, getting an appointment for an abortion at a hospital was very difficult: the hospitals said it's not an emergency case, abortion is not an emergency, so, it's not necessary, so stay out of the hospitals (HP 8).
According to one of the HPs, whether a hospital performed abortions during the pandemic depended on how the hospital leaders viewed abortion: there are many clinic head of departments, mostly male, who will say 'no I don't do abortions in general' or 'in a pandemic I don't do abortions' (HP 1).
However, HP5 and HP7 believed that the counselling centers and the abortion clinics that categorized abortion services as essential health services could potentially have performed more counselling sessions and abortions than normal during the pandemic because other non-essential health services were postponed or stopped.
Another challenge mentioned by HP2 and HP4 was, that due to the COVID-19 restrictions in the clinics, women were not allowed to bring a friend, partner, or family member to the procedure. HP4 described an incident where the companion of a woman having an abortion was turned away at the door and the woman cried "I cannot do this alone" (HP 4).
In response to the pandemic, the Government allowed the counselling centers to provide counselling via telemedicine. According to one of the counsellors (HP 4), the counselling centers would not have started offering telephone and video counselling had it not been for the pandemic. Almost all the HPs maintained that providing counselling via telephone or video calling improved access for many women. Some benefits of telephone and video counselling include not needing to find childcare, take many hours off from work, or travel to the counselling center (HP 4, 5).
In addition, according to some HPs, many women found telephone and video counselling more intimate and private than face-to-face counselling. Many women were more comfortable talking about this "private and somehow shameful" situation because they felt safe at home (HP 5). One counsellor said: I think that [telephone and video counselling] created for some women better access, easier access, and if we offer both [telephone and video, and face-to-face counselling], then that will give more possibilities to the women (HP 4).

Axes of difference
From their experience, HPs identified seven factors that influenced women's access to abortion services during the pandemic: socioeconomic status, disability, language, place of residence, caring responsibilities, home environment, and refugee and immigrant status. HP4 summarized the impact of the pandemic on access to abortion services as follows: In the middle, there's a big number of women for whom it did not change that much […] And on the edges, there are some who have benefitted, and some who really had a lot of problems […] (HP 4).

Socioeconomic status
Several of the HPs mentioned that financial constraint is a barrier to access, which has been exacerbated by the pandemic. Covering the abortion costs was already difficult for women with low or no income before the pandemic but women "had more struggle finding the money to pay for the abortion during the pandemic because they lost their jobs" (HP). The application process for the federal state to cover the costs of the abortion also became more complicated and/or took longer, delaying access to abortion. Applying in person became more difficult because not all the offices were open for in-person appointments, sending the application form in the post took longer, and applying electronically was not always possible.
Lacking the necessary equipment to access telephone or video counselling was another problem, according to two of the counsellors. Women without access to information and communication technologies cannot participate in telephone or video counselling: If you don't have a home office, and if you don't have a device or a good mobile phone, or a notebook [tablet] or whatever, then it was more complicated to reach us [the counselling center] (HP 5).
Not being able to afford PPE was also a barrier to accessing abortion services for some, particularly for those that relied on public transportation when wearing FFP2 masks became mandatory on any form of public transport.
If you don't have the money to buy this [masks], how can you come to the counselling center? (HP 5).

Disability
According to HP4 and HP5, it was harder for women with disabilities to access counseling during the pandemic. For many women with disabilities, telephone and video counselling was complicated or simply not an option: I cannot do video counselling with someone who is blind […] I cannot invite someone who cannot hear with someone who translates with video, it's too complicated.
[…] for those people the access is really bad (HP 4).
Many women with disabilities may therefore only be able to attend face-to-face counselling. As mentioned previously, some counselling centers completely stopped offering face-to-face counselling in the first few months of the pandemic, and many had reduced capacity throughout the pandemic, which resulted in longer waiting times (HP 4, 5).

Language
According to all the counsellors, accessing counselling was more difficult for women not speaking German during the pandemic. It was particularly difficult for women that neither spoke German nor English and therefore needed a translator (HP 5). According to HP5, the social distancing measures made finding someone to translate face-to-face counselling sessions difficult because most of the translators are "community people" or friends of the women, not professional translators (HP 5). Many of the counselling centers also stopped offering face-to-face counselling for some periods during the pandemic and translating through telephone or video is more complicated than in person (HP 4, 5). Lastly, according to HP5, most of the information on the COVID-19 measures from their counselling center and the City of Berlin was in German, making it difficult for non-German speaking people to access information about abortion services and COVID-19 measures.

Place of residence
According to all the HPs, accessing abortion services is more challenging in certain areas of Germany due to the geographical distribution of counselling centers and clinics and hospitals that perform abortions. Many of the HPs mentioned that women in Bavaria and other rural regions, including North Rhine-Westphalia, Rhineland-Palatine, Baden-Württemberg, and the northern parts of Hesse, often need to travel for 100 km or more for abortion care. Since some health facilities stopped providing abortion services during the pandemic, women might have had to travel even further. The lack of childcare during the pandemic also made it more difficult to travel far for abortion. According to HP8, the combination of lack of childcare and distance to the nearest center or clinic was a common barrier to abortion access in rural areas. Three of the HPs (HP 3, 7, 8) said that the abortion providers in the rural areas of Germany are older, which means they are at higher risk of developing serious illness from COVID-19. Some may therefore have chosen to retire or work less during the pandemic, as mentioned previously.
The pandemic also impacted women travelling to or from Germany for abortion (HP 4,5,7,8). Due to restrictive abortion laws, many Polish women travel to Germany to access abortion, while many women in Germany travel to the Netherlands (HP 5, 7). According to HP5 and HP7, cross-border travel for abortion was much more complicated during the pandemic due to border closures and travel bans. HP7 described women travelling to and from Germany for abortion as "really, really vulnerable" during the pandemic (HP 7).

Caring responsibilities
According to several HPs, women with children or relatives to care for, found it challenging to access abortion services during the pandemic. With schools and kindergartens closed, many women had to stay at home with their children.
[…] she was telling me that […] the next possibility for an abortion would be four hours [away] and she had two children, and she didn't know how to manage that. To be that far away. Because it was four hours, so it was not a […] day trip. So, she couldn't get back. And then she said, 'I can't do it' (HP 7).
HP7 therefore identified single mothers and women who "don't have a lot of social support" as particularly vulnerable during the pandemic (HP 7). Furthermore, many could not ask their friends for help with childcare due to isolation, quarantine, and social distancing measures, or because their friends were afraid of catching COVID-19 or busy taking care of their own children (HP7). Many women could not leave the house because their children were in quarantine, while others avoided going to the hospital and other places with "a high risk of catching COVID-19" because their partners or relatives were in a high-risk group (HP 6).

Home environment
Accessing abortion services and keeping the abortion secret was difficult for women living with unsupportive or abusive partners or relatives during the pandemic (HP 5, 6, 8). HP8 said that there were a lot of women experiencing domestic violence who could not get out of the house to access abortion services during the lockdowns. According to HP6, keeping the abortion secret from abusive partners was "way more difficult" during the lockdowns because everyone was at home "24/7" (HP 6). Some women tried to keep the telephone or video counselling secret from their cohabitants by calling the counsellors from outside their homes: […] there were some strange situations, women doing video chats via mobile from shopping malls, in cars, because they don't have a safe space at home. (HP 5).
During the pandemic, many women told WoW that they feared being "kicked out of the house" if their family found out about the abortion (HP 6).

Refugee and immigrant status
The pandemic worsened the pre-existing barriers to abortion access faced by refugees and undocumented migrants (HP 3, 7, 8). The outpatient clinics that care for undocumented migrants and uninsured people struggled to stay open because most of the providers who worked there were retired professionals and at higher risk of developing serious illness from COVID-19 (HP 8). According to HP8, "all these people without papers had immense problems, tremendous problems" (HP 8). HP7 was contacted by a refugee woman living in a reception facility for refugees during the pandemic. She was in COVID-19 quarantine and therefore could not leave the facility to get an abortion. According to HP7, "she was really locked in for two weeks so we couldn't care for her in the end" (HP 7). By the time she got out of quarantine, it was too late to have a medical abortion, but luckily, she managed to have a surgical abortion (HP 7).

Discussion
We assess our data against the concepts of structural violence and intersectionality. Structural violence is explained as the force of social, economic, and health-related policies that act as an invisible force influencing peoples' lives and health in negative ways [12]. People who are already vulnerable are more likely to become affected by such policies. De Maio and Ansell argue that "structural violence" has insufficient explanatory power alone and should be applied in combination with a larger theoretical framework [12]. Intersectionality-theory helps explain how multiple overlapping forms of disadvantage may intersect to obstruct access to health services, in this case to abortion services in Germany during the COVID 19 pandemic. This study shows that access to abortion was made more difficult during the pandemic and draws attention to the ways in which access was particularly difficult for some women due to socio-cultural circumstances in their lives.
An example of structural violence is the Government's failure to recognize abortion services as essential healthcare. With this inaction, the Government rendered this service less legitimate and thereby both exacerbated existing stigmatization [26] and reduced access. Stigma is in itself recognized as a factor that negatively impacts access [6]. Women may avoid seeking services in the formal health system because of fear of stigma [27], which was the case for around 38% of the respondents in this study. Both lack of access and increased stigmatization can be perceived as a form of structural violence.
This study found that some clinics and hospitals in Germany stopped providing abortion services during the pandemic, consistent with a study by Bojovic et al. [9]. Similarly, a study in France found that many women could not find health facilities that still performed abortions during the pandemic [28]. Furthermore, some women had to travel further for abortion. Romanis and Parsons found that distance and lack of transportation was a major problem in countries with vast rural areas both before and during the pandemic, particularly for women with low or no income and for those who rely on public transportation [29]. Accessing abortion services was therefore most likely a significant issue for low-income women in rural areas of Germany during the pandemic.
Due to gender norms of society and family traditions, women may to a significant degree be assigned the role as carers, which made access difficult for three groups of women during the pandemic: women that travel far to access abortion services; women who care for someone in a high-risk group; and women bringing up children alone or with limited social support. These findings correspond to previous studies [28,29].
A study by Killinger et al. found that financial constraint was a major barrier to accessing abortion services in the formal health system in Germany before the pandemic [5]. This study found that the pandemic exacerbated the pre-existing financial barriers to abortion access. More than one third (36%) of the respondents in this study reported that costs was a barrier to accessing abortion services in the formal health system. Furthermore, the qualitative data showed that some women lost their jobs during the pandemic and therefore could not afford an abortion. Research supports that the pandemic and lockdown measures increased unemployment in Germany [30].
Our findings revealed that secrecy and privacy issues were central to accessing abortion services. Almost half of the respondents reported needing to keep the abortion secret from their partner or family. This speaks to abortion stigma and threats to reproductive autonomy. The qualitative results indicated that women who experience violence or live in a controlling environment faced barriers to accessing abortion services during the pandemic, consistent with a previous study from France [28]. Threats to reproductive autonomy can arise from interpersonal relationships, particularly in association with intimate partner violence, as well as from health systems and other structural sources, including COVID-19 restrictions [31].
The qualitative results also illustrate how refugee status affects access to abortion services in Germany, and most likely in other countries as well, during the pandemic. In Germany, abortion services were delayed for women who experienced symptoms or tested positive for COVID-19, and the risk of transmission is particularly high in reception facilities, as the refugees are living in confined, shared spaces [10,32]. Refugee status therefore renders women more vulnerable to COVID-19 infection, which impacts access to abortion services.
The qualitative data also indicate that women had abortions later in their pregnancies, possibly due to the lack of timely access to abortion services. One study found that getting an appointment for an abortion could take up to two weeks in Germany during the pandemic [9]. Delayed access to abortion services may thus have caused some women to exceed the legal gestational age limit for medical or surgical abortion in Germany during the pandemic. Abortions later in the pregnancy are often stigmatized. Furthermore, previous studies confirm the link between later presentation for abortion and social disadvantage, including experiences of reproductive coercion [33].
On the bright side, the qualitative results indicate that the provision of counselling via telephone or video alleviated some of the barriers women in Germany faced before the pandemic. However, women with visual or hearing impairments or other disabilities who cannot attend telephone or video counselling were further marginalized. According to Romanis and Parsons, women with disabilities are more likely to experience barriers resulting from COVID-19 measures [29], which supports the qualitative results in this study.
Finally, the Government did not officially allow the use of telemedicine abortion within the formal health system in Germany. In contrast, the UK Government responded to the pandemic by allowing the use of telemedicine abortion within the formal health system [9]. According to Aiken et al., the response of the UK Government resulted in a significant decrease in requests for telemedicine abortion through WoW from the UK [8]. Considering the findings of Aiken et al. [8], providing telemedicine abortion within the formal health system may reduce barriers to abortion services in Germany.

Study strengths and limitations
One of the main limitations of this study is the lack of qualitative data on the experiences and perceptions of women seeking abortion services. Another limitation is the qualitative sample size, as well as the limited number of participants from outside Berlin. Recruiting more HPs was made difficult due to abortion stigma and the significant work-related stress caused by the pandemic. The quantitative data is also not representative of all women seeking abortion services in Germany, and may involve selection bias, since it consists of people seeking abortion services outside the formal heath system. Another limitation is that the dataset consists of self-reported data. We acknowledge that our findings cannot make an overall assessment of abortion service delivery in Germany during the pandemic. Future quantitative and qualitative studies could address these limitations.
To the best of our knowledge, this study is one of the first attempts to explore the impact of the COVID-19 pandemic on access to abortion services in Germany by combining patient and provider perspectives and experiences. This study therefore has potential to inform HPs and policymakers on barriers to accessing abortion services and the actions needed to ensure access to abortion services for all, both in Germany and in other countries.

Conclusions
This study provided insight into the impact of the COVID-19 pandemic on access to abortion services in Germany. The pandemic exacerbated existing and created new barriers to accessing abortion. The pandemic affected the provision of abortion services as well as the circumstances of the women seeking abortion. As a result, accessing abortion services was more difficult for many women in Germany during the pandemic, especially for women experiencing multiple and overlapping forms of discrimination.
The pandemic lays bare the existing fault lines in the German healthcare system. Significant legal and policy barriers to accessing abortion services remain in Germany. To improve access to abortion services and ensure SRHR for all, the German Federal Government should recognize abortion as an essential health service and eliminate all other discriminatory abortion laws and policies, including the requirement for pre-abortion counselling. Considering the findings of the present and previous studies, the German Federal Government should also allow the provision of telemedicine abortion within the formal health system during and beyond the COVID-19 pandemic.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest
Coauthors HR and BT declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Coauthors RG and HA work for or are affiliated with WoW.