Abortion is a common and commonly needed healthcare procedure. Approximately 121 million unintended or ill-timed pregnancies occur each year globally; more than half end in abortion (Bearak et al. 2020). Abortion access is of importance to the field of global health given that unintended pregnancy rates are highest and abortions are less safe in countries with more abortion restrictions, while unintended pregnancy rates are lower and abortions generally safer in countries with greater legal access (Bearak et al. 2020; Ostrach 2013; 2016). Regardless of legality, abortion demand and rates are similar globally. However, the proportion of unintended pregnancies that end in abortion in more restrictive countries has increased in recent years (Bearak et al. 2020). What varies most is not abortion demand but rather legality, safety and access. All of these factors are intimately linked to the degree of societal stigma surrounding abortion, with important implications for global reproductive health outcomes (Ostrach 2016).
Publicly Funded Abortion in Catalunya
Catalunya is an autonomous European nation bordering southeast France and northeast Spain and long occupied by Spain (Ostrach 2017; 2020; Crow 1985). In Catalunya, legal access to abortion expanded under a 2010 Spanish law liberalising the gestational limits and circumstances under which a pregnant person could obtain an abortion on request or with authorisation from a healthcare provider (i.e. a physician or psychiatrist) (Alan Guttmacher Institute 2010; Ostrach 2017). From 1985 to 2010, abortion was legal in Spain and Catalunya, but only with authorisation from a physician or psychiatrist on the grounds listed above and was not broadly included in the publicly funded health system. The 2010 law liberalising abortion access covered all procedures in the first trimester—considered to be up to 14 weeks from the last menstrual period [LMP]) —and allowed procedures in the second trimester—up to 22 weeks from LMP—if authorised based on foetal anomaly or threats to the pregnant person's physical or mental wellbeing, including financial stress or relationship precarity (European Parliament 2016). The 2010 liberalisation also created an avenue to include abortion coverage through the publicly funded health system if the pregnant person obtains required referral paperwork from a public health system centre (Ostrach 2017; Serret and Pairó 2018).
A study conducted prior to abortion decriminalisation in Spain found rates of women travelling from Spanish regions to England, Wales and the Netherlands for abortion increased consistently from 1974 to 1985 (Peiró et al. 2001). After passage of a 1985 Spanish law allowing abortion under certain circumstances (with authorisation), no effect was observed on abortion rates overall; instead, research suggested more women obtained abortions in Spain, but fewer in England, Wales and the Netherlands (Peiró et al. 2001). Investigators concluded decriminalisation ‘[made] abortion available locally and … [thereby reduced] inequalities implied by lack of access to proper health care services’ (Peiró et al. 2001: 1).
Abortion rates in Catalunya declined from 2008 to 2014, despite having increased steadily in the preceding decade (Serret and Pairó 2018). Further, these data showed that migrants (e.g. individuals not born in Catalunya) had a higher abortion rate than did Catalans, consistent with other studies on abortion in Catalunya (de Ramales et al. 2006; Gispert Magarolas et al. 2008; Ostrach 2013; Serret and Pairó 2018). Though financial precarity is often cited as a reason for seeking an abortion, especially during and in the years soon after the global economic recession beginning in 2007–2008 (Perez et al. 2019; Ostrach 2017), prior research suggests the abortion rate in Catalunya remained the same just before and at the time considered the height of the economic crisis (‘La Crisis’) (Perez et al. 2019).
As previously published in this journal (Ostrach 2020), by early 2016 the actual process by which residents in Catalunya could navigate the public health system to obtain publicly funded abortions did not match either the existing law or official policies. Many who attempted to obtain abortions encountered delays and disinformation. Such delays and obstacles were often worse for migrants, low-income residents and other marginalised people. This was the case soon after the 2010 abortion liberalisation (Ostrach 2020), yet even years later notable delays and lack of clear and accurate information provided by health system staff persisted. Delays in receipt of safe, high-quality, legal abortions are well demonstrated to increase risks of negative physical, mental health and even socio-economic sequelae (Ostrach 2016; Foster et al. 2018a, 2018b; Mayi-Tsonga et al. 2009). Abortion delays can also increase the need for those seeking care to travel and navigate other logistical obstacles, such as childcare, time off work and related economic burdens (Ostrach and Cheyney 2014; Foster et al. 2008; Harries et al. 2007).
Objectives
In this article, I report a detailed comparison from two phases of rapid ethnographic assessment (REA) (Utarini et al. 2005) conducted six months apart at an abortion clinic contracted with the Catalan public health system in 2016. This comparison also functions as a follow-up to the first phase of REA previously published in this journal (Ostrach 2020). The objective is to (1) compare the first wave to the second to determine whether delays in obtaining required public health system vouchers for publicly funded abortions persisted; (2) if so, if such delays continued to appear more pronounced for migrants and other marginalised people; and (3) to identify any other changes or continuities between the first and second REA. Both waves of data collection reported here occurred at the invitation of and in collaboration with a full-spectrum abortion clinic contracted with the Catalan public health system (Ostrach 2017; 2018; 2020). This analysis builds on a larger research project that first began in 2012.
Methods
Research Site and Access
Each wave of data collection took place at a Barcelona clinic contracted with the Catalan public health system to provide publicly funded abortion care to patients who obtain a required referral voucher from a health system centre. This site is referred to in earlier publications, and in this article, as the ‘Today Clinic’ (Ostrach 2017; 2018; 2020). I have been granted long-term, periodic access to this site on an ongoing basis and have been invited to come for research visits at various intervals. I received input from the clinic director and other key informants on staff about salient topics for ethnographic investigation. Protocols for each phase of REA were reviewed and found exempt from further review by the Institutional Review Board at Boston University School of Medicine.
Sampling and Recruitment
In both phases I engaged in opportunistic and convenience sampling to recruit patients. When arriving at the clinic, all patients received a flyer describing the study and stating that they would receive the same quality of care whether they elected to participate or not. In most cases I was able to introduce myself to patients during their intake or discharge and asked if they would be interested in talking with me after their appointment. For all observations in the clinic, staff asked the patient's permission prior to my being in the room. All clinic staff consented to my being present and observing while they worked, for participant observation. Both phases of REA were unfunded; no incentives were offered.
Research Focus Areas
As previously reported in this journal, of interest in the initial phase of REA conducted in January 2016 were ongoing longitudinal (2012–2016) assessment questions—including how many visits to public health system centres and how much of a wait patients experienced to obtain the referral voucher required for an abortion to be paid for by the health system—and any inequities in wait times or number of visits for marginalised populations. Of interest in the subsequent phase of REA in June 2016 were the longitudinal assessment questions listed above and especially whether any changes could be observed.
Data Collection and Analysis
Data collection in both phases consisted of informal interviews conducted on-site at the clinic after a patient's medical appointment was complete and/or off-site at another time, depending on the participant's preference. With patient consent, I engaged in participant observation as a clinic volunteer multiple days per week, embedded primarily in the Medical Social Work office and other patient care spaces. Participant observation involves being present in the research environment in a deeply embedded way, engaging in activities that allow for direct observation of the experiences of interest in the research question and aids in research by providing context and facilitating triangulation of data collected through more structured methods (Ostrach 2017; Bernard 2011). For this article, I calculated descriptive statistics using quantitative data compiled from verbatim transcripts of informal interviews, detailed notes from informal interviews and detailed fieldnotes taken during participant observation.
Results
As described elsewhere (Ostrach 2020), the earlier period of REA in January 2016 yielded informal interviews (n = 28 patients). Descriptive statistics and an analysis of patient interviews focused on disparities in access for migrant and economically marginalised populations were earlier reported in this journal (Ostrach 2020). Here I report a comparison and follow-up data from the subsequent June 2016 REA (n = 31 patients).
In June 2016 I closely observed intakes and/or other portions of contracted clinic appointments with at least 29 patients, taking detailed notes. Two patients elected to participate in informal, audio-recorded interviews while still at the clinic following the conclusion of their appointment. Together this yielded a total estimated June REA patient n = 31. I also conducted key informant interviews with five clinic staff—the director, staff psychologist, director of social work, receptionist, and a longtime employee trained in social work who conducts patient intakes and discharges and also assists with medical assistant tasks as needed.
Participant Characteristics
Sample characteristics are presented in Table 1. In the second phase of REA, participants ranged in age from 16 to 44 years old. The average age of observed patients was 28; the median was 23. Among patients for whom trimester of pregnancy was known, 31% were in the first trimester and 69% were in the second. Average gestation among patients for whom specific weeks of pregnancy was observed was between 16 and 17 weeks. Range of gestations was six to just over 22 weeks (very edge of the legal limit). Of patients for whom nationality and residency status were known, 45% were migrants. The average time migrants had been in Catalunya was 10 years, with a range of 6–13 years.
Comparison of participant characteristics
Public Health System Experiences
Seventy-one percent of observed patients in the second phase of REA obtained the health system referral voucher required for an abortion to be covered at the contracted clinic. In the second phase of REA some participants reported they were required to go to a different centre than their usual neighbourhood health system clinic after initially being denied a voucher. The average number of required visits to a public health system centre necessary to receive a referral voucher was 2.5. In the second phase of REA the median number was two visits and the range was 1–5. In the second phase of REA the range of waiting times was less than 1 week up to 3 weeks (see Table 2).
Comparison of public health system experiences
Difficulty Obtaining Vouchers
In the second phase of REA, while just 10% of observed patients reported it was hard to obtain the referral voucher, 19% explicitly said it was easy. In the second phase of REA, 10% of patients ended up paying privately for an abortion that could have been paid for by the public health system if they had been told about coverage or if appropriate paperwork had been prepared. Yet another 6% of observed patients in the second phase sought a referral voucher from a public health system centre where staff refused to provide a voucher, or completed it incorrectly thereby rendering it invalid.
Migrants’ Experiences
Of participants in the second phase of REA for whom migration status was known, those not born in Catalunya waited on average three times as long as Catalans to obtain the required referral voucher. However, migrants in the second phase of REA did not have to make more visits to a neighbourhood health system centre to obtain a voucher than Catalans. This appeared to be only a slight improvement from the first phase of REA, where substantially more migrants (nearly 70%) had to make more visits and/or wait longer than they should have had to by policy to receive the voucher. This was in contrast to less than half of locally born participants (41%) who had to make more visits and/or wait longer.
Discussion
Findings from these two phases of REA further illuminate and expand on previous longitudinal research about patient and clinic staff experiences of seeking and providing publicly funded abortions in Catalunya. It is also possible to assess short-term changes or continuity evident in the experiences of participants between the first and second phases.
Participants in both phases of REA were of similar ages; the range of ages was slightly wider and the average age just slightly younger in the second phase. Nearly the same proportion of participants in both phases were migrants (not born in Catalunya). Average length and ranges of time that migrants had been in Catalunya were very similar, with a slightly wider range in the first phase of REA. In both phases, migrants had come from many countries.
The proportion of patients in first versus second trimesters of pregnancy was markedly different between the two phases of REA, with nearly two-thirds of patients interviewed in early 2016 being in the first trimester. This was true for just under a third of patients for whom gestation was known in the second REA phase. The shift in trimester of pregnancy was also reflected in a later average gestation in the second phase and a shift from average gestation in the first trimester to the second trimester. This later average gestation in the second phase may have reflected what appeared to be an increasing proportion of public health system patients who reported being encouraged or automatically scheduled for pharmacologic abortions directly onsite at a public health system centre rather than being offered a referral voucher for a contracted clinic. In such a scenario, where the pharmacologic abortion occurs in the first trimester, fewer first trimester patients would reach the contracted clinic, leaving a greater proportion of patients overall being those in the second trimester.
It is also worth noting that in the second phase of REA the clinic was fast approaching a quota set by the health system for the total number of vouchers it could accept under its annual contract with the health system. Thus, clinic schedulers were beginning to encourage patients, when calling to schedule an appointment, to pay out of pocket if they could do so (in order to conserve voucher appointments for the remainder of the year). Patients in the first trimester may have been more likely to be able to pay out of pocket given the lower cost of an early abortion, rather than going to their neighbourhood health system centre to seek a public funding referral voucher.
The range in the number of visits necessary to obtain a public health system referral voucher to have their abortions paid for by the public health system was the same in both phases of REA (1–5 visits). Though participants in the second phase of REA had later average gestations, participants in the second phase of REA reported slightly shorter average wait times to receive vouchers (less than 1 week to 4 weeks vs less than 1 week to 3 weeks). The slightly shorter average wait times in the second phase may have reflected the narrower range of wait times (1.2 weeks vs 1 week). Though the range in number of visits people reported as necessary to obtain vouchers was the same in both phases (1–5 visits), participants in the second phase had to make just slightly more visits, on average, to obtain a voucher (2.5 versus 1.9) (Ostrach 2020).
Delays and wait times to obtain a referral voucher did appear to have a greater impact on migrants, as in the earlier phase of REA and consistent with an earlier, longer-term ethnographic study in the same clinic (Ostrach 2020). As such health system delays are known to increase health and social risks, and compound and be compounded by other delay factors likely to be more burdensome for already marginalised populations such as migrants, people in poverty and people travelling greater distances for care, even small disparities in delays for migrants and other marginalised populations are troubling and deserving of attention.
On balance, it does not appear that, between the first and second phases of REA, obtaining a health system referral voucher for a publicly funded abortion became more difficult overall, though based on participant observation it appeared to be easier for people earlier in pregnancy and more difficult for people whose pregnancies were further along. It is also notable that in the second phase of REA, fully six years after passage of the abortion liberalisation laws and integration of abortion into the public health system, nearly one in five participants was denied a voucher, or received an invalid voucher.
The shift toward later gestations in patients arriving at the contracted clinic with referral vouchers, in the second phase, was the most notable difference within such a short period of time. As mentioned above, the most obvious explanation is the widely reported increase in public health system staff automatically scheduling patients in the first trimester for a pharmacologic abortion at a main health system centre unless they request a voucher for a contracted clinic, and some public health system staff otherwise pressuring first trimester patients who might otherwise have gone to a contracted clinic for an instrumental abortion to have a pharmacologic abortion. As clinic staff were beginning to remark on this trend in the first phase of REA, and by the second phase expressed concerns about the ability of the clinic to stay open in part due to reduced reimbursement quotas related to the public health system attempting to keep more abortion patients at main centres, it was unsurprising that a greater proportion of patients who would have been ineligible for a pharmacologic abortion were still seen at the contracted clinic.
Limitations
As a qualitative, observational, rapid ethnographic study, sample sizes were small enough that differences noted between the first and second phases of REA do not constitute statistical significance. All quantitative calculations are descriptive and offered for the sake of comparison. Some data points could not be compared, as they were not collected in both phases of REA: for example, in the first phase of REA an inclusion criteria was that all participants have obtained the required public health system referral voucher. Therefore, the percentage of all abortion patients at the contracted clinic who had obtained a voucher, observed in phase 2 of REA, could not be compared with the same indicator in phase 1. As with other clinic-based studies, these are the experiences and perspectives of those patients who successfully made it to the contracted clinic, in most cases after obtaining the required referral voucher. Therefore, absent from this account are the experiences of people who sought publicly funded abortions but were unable to obtain a voucher and lacked funds to pay privately at this clinic. In addition, the perspectives of public health system patients who underwent a pharmacologic abortion at a public health system centre are not represented here, although many patients I interviewed and spoke with during participant observation reported having had a pharmacologic abortion previously, within the public health system, and intentionally requested a referral voucher for the contracted clinic to avoid repeating that experience. These data were collected prior to the current COVID-19 pandemic, so I cannot speak directly to the impact of COVID-19 on the Today Clinic or its staff or patients, though I remain in contact with key informants in the clinic who reported the clinic remained open throughout city-wide quarantine periods and curfews.
Conclusion
Based on two periods of REA, conducted six months apart in 2016 at a clinic where I have conducted longitudinal research about experiences of obtaining and providing publicly funded abortion in Catalunya since 2012, I found that certain earlier trends in the number of visits and waiting times to obtain the required public health system vouchers required for publicly funded abortions persisted, namely migrants experiencing longer wait times to obtain the voucher and some people being unable to obtain the voucher at all. Differences were observed in the gestations at which publicly funded abortions were obtained at the clinic where I collected data. Namely, by the second phase of REA, more patients were arriving for abortion care at the contracted clinic much further along in their pregnancies.
As a result of consultation and collaboration with key informants in the research site, data collection in both phases of REA focused on patients’ experiences obtaining the voucher required by the Catalan public health system for abortions to be paid for at a contracted clinic. Comparing these two phases of REA conducted within a short time period, six years after abortion laws were liberalised and abortion care ostensibly incorporated into the Catalan health system, many patients seeking publicly funded abortion nevertheless continued to encounter delays in their ability to obtain the referral voucher required for an abortion to be paid for at a contracted clinic.
To the extent that unnecessary repeat visits and delays in obtaining required referral vouchers persisted, these represented a continuation of the civilised oppression (Harvey 1999) long evident in the Catalan public health system (Ostrach 2017; Biglia and Olivella-Quintana 2014). With the effects of COVID-19 further straining the Catalan health system (Carney and Ostrach 2020; Mena-Tudela et al. 2022), the necessity of monitoring how health system policies are truly enacted and whether access on paper translates into accessible care is ever more important. Unsafe abortion is a recognised global health threat, made more likely where access is restricted or complicated. To ensure safe and accessible abortion, settings where the procedure is legal and publicly funded must also guarantee practical access.
Acknowledgements
With gratitude to the staff and patients of the ‘Today Clinic’ for welcoming me. As always, thank you to Marion Malcolm and Stefan for making this fieldwork possible. Huge thanks to Dr Ashley R. Houston and Stefan Ostrach for co-writing sessions and for insightful feedback on an early draft of this article. Thanks also to the copyeditor and production team at Anthropology in Action and Berghahn Journals.
References
Alan Guttmacher Institute (2010), ‘Spain Expands Legal Access to Abortion’, https://www.guttmacher.org/media/inthenews/2010/03/03/index.html.
Bearak, J., A. Popinchalk, B. Ganatra, A.-B. Moller, Ö. Tunçalp, C. Beavin, L. Kwok and L. Alkema (2020), ‘Unintended Pregnancy and Abortion by Income, Region, and the Legal Status of Abortion: Estimates from a Comprehensive Model for 1990–2019’, The Lancet Global Health 8, no. 9: e1152–e1161.
Bernard, H. R. (2011), Research Methods in Anthropology: Qualitative and Quantitative Approaches (Lanham: Rowman Altamira).
Biglia, B. and M. Olivella-Quintana (2014), ‘Evolution and Involution in the Sexual and Reproductive Health Services in Catalonia (Spain)’, Women's Studies International Forum 47: 309–316, http://www.sciencedirect.com/science/article/pii/S0277539514000338.
Carney, Megan A., and Bayla Ostrach. 2020. “Austerity, Not COVID-19, Strains National Healthcare Systems.” Somatosphere, April 28, 2020. http://somatosphere.net/2020/austerity.html/.
Crow, J. A. (1985), Spain, the Root and the Flower: An Interpretation of Spain and Spanish People (Berkeley: University of California Press).
European Parliament (2016), ‘Citizens’ Rights & Constitutional Affairs: Gender Equality Policies in Spain, Update’, Gender Equality. Directorate General for Internal Policies, https://www.europarl.europa.eu/RegData/etudes/STUD/2016/583112/IPOL_STU(2016)583112_EN.pdf.
Foster, D. G., R. A. Jackson, K. Cosby, T. A. Weitz, P. D. Darney and E. A. Drey (2008), ‘Predictors of Delay in Each Step Leading to an Abortion’, Contraception 77, no. 4: 289–293, https://doi.org/10.1016/j.contraception.2007.10.010.
Foster, D. G., M. A. Biggs, S. Raifman, J. Gipson, K. Kimport and C. H. Rocca(2018a), ‘Comparison of Health, Development, Maternal Bonding, and Poverty among Children Born after Denial of Abortion vs after Pregnancies Subsequent to an Abortion’, JAMA Pediatrics 172, no. 11: 1053–1060.
Foster, D. G., M. A. Biggs, L. Ralph, C. Gerdts, S. Roberts and M. M. Glymour (2018b), ‘Socioeconomic Outcomes of Women Who Receive and Women Who Are Denied Wanted Abortions in the United States’, American Journal of Public Health 108, no. 3: 407–413.
Gispert Magarolas, R., G. Clot-Razquin, M. del Mar Torné, R. Bosser-Giralt and A. Freitas-Ramírez (2008), ‘Diferencias en el Perfil Reproductivo de Mujeres Autóctonas e Inmigrantes Residentes en Cataluña’, Gaceta Sanitaria 22, no. 6: 574–577, https://doi.org/10.1016/S0213-9111(08)75356-1.
Harries, J., P. Orner, M. Gabriel and E. Mitchell (2007), ‘Delays in Seeking an Abortion until the Second Trimester: A Qualitative Study in South Africa’, Reproductive Health 4, no. 1: 7.
Harvey, J. (1999), Civilized Oppression (Cambridge: Cambridge University Press).
Mayi-Tsonga, S., L. Oksana, I. Ndombi, T. Diallo, M. H. De Sousa and A. Faúndes (2009), ‘Delay in the Provision of Adequate Care to Women Who Died from Abortion-Related Complications in the Principal Maternity Hospital of Gabon’, Reproductive Health Matters 17, no. 34: 65–70.
Mena-Tudela, D., S. Iglesias-Casas, A. Cervera-Gasch, L. Andreu-Pejó, V. M. González-Chordá, and M. J. Valero-Chillerón(2022), ‘Breastfeeding and Obstetric Violence during the SARS-CoV-2 Pandemic in Spain: Maternal Perceptions’, International Journal of Environmental Research and Public Health 19, no. 23: 15737.
Ostrach, Bayla. 2013. “‘Yo No Sabía...’—Immigrant Women's Use of National Health Systems for Reproductive and Abortion Care.” Journal of Immigrant and Minority Health 15 (2): 262–72.
Ostrach, Bayla. 2016. “This Tangled Web of Reproductive Morbidity Risk: Abortion Stigma, Safety and Legality.” Frontiers in Women's Health 1 (2): online.
Ostrach, Bayla. 2017. Health Policy in a Time of Crisis: Abortion, Austerity, and Access. Routledge.
Ostrach, Bayla. 2018. “Social Movements, Policy Change, and Abortion Access in Catalunya.” Anthropology Now 10 (2): 1–11.
Ostrach, Bayla. 2020. “Publicly Funded Abortion and Marginalised People's Experiences in Catalunya: A Longitudinal, Comparative Study.” Anthropology in Action 27 (1): 24–34.
Ostrach, Bayla, and Melissa Cheyney. 2014. “Navigating Social and Institutional Obstacles: Low-Income Women Seeking Abortion.” Qualitative Health Research 24 (7): 1006–17.
Peiró, R., C. Colomer, C. Alvarez-Dardet and J. R. Ashton (2001), ‘Does the Liberalisation of Abortion Laws Increase the Number of Abortions? The Case Study of Spain’, European Journal of Public Health 11, no. 2: 190–194, https://doi.org/10.1093/eurpub/11.2.190.
Perez, G., M. Gotsens, C. Cevallos-García, M. F. Domínguez-Berjón and IMCrisis Project group (2019), ‘The Impact of the Economic Recession on Inequalities in Induced Abortion in the Main Cities of Spain’, European Journal of Public Health 29, no. 2: 279–281.
Ramales, A. Llácer Gil de, C. Morales Martín, S. Castillo Rodríguez, L. Mazarrasa Alvear and M. L. Martínez Blanco (2006), ‘El Aborto en las Mujeres Inmigrantes. Una Perspectiva Desde los Profesionales Sociosanitarios que Atienden la Demanda en Madrid [Abortion in Immigrant Women: The Perspective of the Social and Health Professionals Who Deal with the Demand in Madrid]’, Index de Enfermería 15, no. 55: 13–17.
Serret, L. F. and M. S. Pairó (2018), ‘La Interrupció Voluntària de l'Embaràs a la Catalunya del Segle XXI’, Papers: Revista de Sociologia 103, no. 1: 75–100.
Utarini, A., A. Winkvist and G. H. Pelto (2005), ‘Appraising Studies in Health Using Rapid Assessment Procedures (RAP): Eleven Critical Criteria’, Human Organization 60, no. 4: 390–400, https://doi.org/10.17730/humo.60.4.3xu3p85amf13avtp.