Implementing medical abortion through telemedicine in Colombia: a qualitative study

Abstract

The non-governmental organisation Profamilia developed and implemented medical abortion through telemedicine in response to the Covid-19 pandemic. This service is now integrated as an alternative to in-person care and available to abortion-seekers across Colombia. Previous research has emphasised bottlenecks in abortion provision, but less is known about implementation processes and experiences. We assessed the feasibility and acceptability of telemedicine for medical abortion from the perspectives of key informants involved in the implementation in Colombia. We conducted 15 in-depth interviews with healthcare professionals, coordinators and support staff implementing telemedicine for medical abortion in the early phase of implementation, between March and October 2021. We analysed the data using the framework method and applied the normalisation process theory in our analysis and interpretation of findings. Our findings show that strong leadership, organisational efforts on pre-implementation training, monitoring and evaluation, and collaboration between diversely skilled and experienced providers are essential for successful implementation. Participants were generally positive towards the use of telemedicine for medical abortion; concerns related to effectiveness, safety and safeguarding existed mainly among providers with less clinical experience. We identified contextual barriers, such as social opposition, regulatory barriers, providers’ unavailability, and poor phone and internet connections in rural areas, which impacted the feasibility of the intervention negatively. In conclusion, to ensure stakeholders' buy-in and for the service to reach all abortion seekers in need, future implementation endeavours must address concerns about safety and effectiveness, and tackle identified contexual barriers.

Plain Language Summary

In telemedicine for medical abortion, all or some components of abortion care, such as initial consultations, home delivery of abortion medication, and post-abortion follow up are provided with the use of telecommunications. Telemedicine for medical abortion has been shown to be a safe and effective form of service delivery.

In this study, we interviewed 15 healthcare providers and staff involved in the implementation of a telemedicine service for medical abortion in Colombia to determine whether they deemed the service to be acceptable and feasible. We found that collaboration between providers of different backgrounds and levels of experience, appropriate training and strong leadership were key factors for successfully implementing the service. However, some healthcare providers, especially those with less clinical experience, were concerned that telemedicine for medical abortion may not be safe and may risk the health and well-being of abortion-seekers. Further, social opposition to abortion, unclear regulation and limited access to technology were identified as barriers that need to be addressed to ensure the service reaches all abortion-seekers in need.

In conclusion, despite contextual barriers and some provider’s concerns about medical safety, telemedicine for medical abortion was viewed as a positive and feasible form of service delivery in Colombia.

Introduction

Telemedicine (TM) is the provision of healthcare services by healthcare workers using telecommunications, where distance is a critical factor.¹ When applied to medical abortion care, TM can be used as a complement to in-clinic care to support the entire medical abortion process or for specific abortion care components.² The use of TM for medical abortion has been shown to be safe and effective in the first trimester and an acceptable alternative to in-person abortion care.^3–5 For low-risk women undergoing medical abortion in early pregnancy, existing evidence supports the use of TM without the need for any in-clinic visits or care.⁶ When abortion services are provided through TM, the need to travel for in-clinic care reduces which may save time and reduce costs. In this way, the intervention can improve timely access to safe self-management, at home and in private.^4,7,8 The use of TM may also mitigate stigma while overcoming geographical, social, or legal barriers to accessing abortion care.^4,7 TM for medical abortion is supported by the World Health Organization (WHO), who recently issued updated abortion care guidelines with a recommendation on the provision of medical abortion, partly or fully, via TM.² By safeguarding and improving access to essential sexual and reproductive healthcare, TM for medical abortion is an important means to upholding individuals’ right to bodily autonomy, privacy, dignity and non-discrimination.⁹

The COVID-19 pandemic significantly impacted access to and utilisation of sexual and reproductive health (SRH) services, including abortion services. Available evidence indicates a decline in access to in-person abortion services, while access to and utilisation of remote abortion services facilitating self-managed abortion (SMA), including medical abortion through telemedicine, increased during this time.^7,10,11 The pandemic led several countries to speed up the implementation of TM for medical abortion to safeguard access, and thereby mitigate the negative effects of some of the measures put in place to prevent the spread of the virus.^12,13 In Latin America, despite recognition of SRH services as essential during the COVID-19 pandemic, abortion was often the exception, with detrimental consequences for the sexual and reproductive health and rights (SRHR) of abortion-seekers, particularly those who are most marginalised. Many countries in this region did not allow TM for medical abortion and maintained requirements of in-person care. This led to a decline in access to and utilisation of legal and safe in-person abortion services and an increase in clandestine and often unsafe abortions.¹⁴

Colombia was among the few countries that allowed the provision of TM abortion services as a response to the pandemic. This decision is partially explained by a history of abortion depenalisation on three legal grounds by Colombia’s Constitutional Court, and the existence of guidelines produced by the Ministry of Health and Social Protection for the provision of abortion services.^15,16 Within this scenario, Colombia pioneered the production of guidelines for providing SRH services through telemedicine and their recognition as priority needs, in line with international human rights standards relating to medical abortion.⁹

In Colombia, TM services for medical abortion were mainly implemented by private and specialised SRH centres but, despite noteworthy efforts by abortion care providers, social, geographical and economic barriers in access and logistical and implementation challenges emerged.^14,16

While a growing body of evidence from Latin America focuses on issues of safety, effectiveness and acceptability of TM for medical abortion among abortion-seekers, less is known about the implementation processes and the factors that affect them through the perspectives of adopters.^17,18 Implementation bottlenecks identified so far include issues of privacy and logistical challenges in medication delivery, but little data exist about the implementation process itself.¹⁶

This study aimed to assess the feasibility and acceptability of TM for medical abortion from the perspective of healthcare providers and staff involved in the coordination and delivery of the service within Profamilia, a non-profit organisation in Colombia which has led the provision and implementation of SRH services in the country for the past 50 years. Specifically, this study aimed to explore the factors that influence the process of implementation of TM as well as healthcare providers’ and staff’s perceptions of the functioning and utility of TM for medical abortion in Colombia.

Methods

Study design

This exploratory qualitative study was conducted using the framework method for the analysis of in-depth interviews with key informants.¹⁹ The study was reported according to an adapted version of the Standards for Reporting Implementation Studies (StaRI) Checklist ²⁰ to include qualitative methodological aspects, for which we followed the Consolidated criteria for reporting qualitative research (COREQ) checklist.²¹

Setting

The study was conducted in collaboration with Profamilia in three of its SRH clinics, in the cities of Apartadó, Bogotá and Medellín. The first versions of the organisation’s TM model for abortion care were implemented in these locations and were therefore deemed suitable as study settings. As an established and trusted organisation with over 50 clinics around the country, Profamilia is the largest provider of SRHR services in Colombia.²² Profamilia began offering a TM abortion service in 2020 during the COVID-19 pandemic, with the purpose of reaching abortion-seekers where and when they need it, including in areas where in-person services are not readily available.

Context

Latin America is a region characterised by a complex scenario for abortion provision, due to the heterogeneity of abortion laws and the pressure of conservative forces on its provision.¹⁵ At the time of data collection in Colombia, abortion had been decriminalised and framed as a human right since 2006. According to the law in place until 2022, women could access a legal abortion in three exceptional cases: when the pregnancy threatens the woman’s life or health, in case of fetal malformation incompatible with life outside the uterus as certified by a doctor, and when the pregnancy results from sexual violence or incest. However, the characteristics of the law were still unknown to a large part of the population and the persistent stigma and taboo around abortion influences access to safe abortion services.²³ These factors combined with poor implementation of the law and a large proportion of conscientious objectors, still force women and girls to seek clandestine and often unsafe abortions.^23,24

Furthermore, the measures taken to control the COVID-19 pandemic in Colombia created new and exacerbated pre-existing barriers to abortion care, to the extent that abortion services were sometimes disrupted. In order to ensure access to safe abortion, the Colombian Ministry of Health and Social Protection prohibited the disruption of such services and updated the regulation related to the provision of abortion by allowing the use of telemedicine.²⁵ This was facilitated by the increasing availability of internet connectivity and technology across the country.²⁶ However, the persisting gender, geographical and wealth gaps in Colombia mean many still do not have access to the internet and technology, and particularly women and girls in rural areas.²⁷

Development and implementation of abortion through telemedicine within Profamilia

The TM service was designed in 2020 by a multi-disciplinary team in Profamilia, including healthcare professionals, but also legal and communications experts. It was tested and implemented in Colombia in 2021, as a response to the mobility restrictions and limited access to safe abortions during the COVID-19 pandemic, particularly for those living in remote areas of the country. The intervention is part of a strategy that seeks to expand the right to safe abortion by facilitating its self-administration and adherence to abortion medications. Abortion-seekers with a gestational age up to 10 weeks are eligible to benefit from the telemedicine service as stated in the guidelines by the Ministry of Health at the time.²⁵

Medical abortion is delivered using TM in three key steps. First, there is an eligibility evaluation via telecommunications followed by a medical prescription of an abortion kit for eligible women. In cases where a woman is uncertain about the dates for her last menstrual period (LMP), or if the health worker identifies a risk for ectopic pregnancy, the woman is recommended to visit one of the Profamilia clinics for further management. This is followed by home-delivery or pick-up at any Profamilia clinic of a kit that contains the necessary medications for medical abortion including analgesia, as well as information brochures on self-administration, a pregnancy test, a condom and, if desired by the abortion-seeker, an oral contraceptive method. Lastly, TM users are followed throughout the entire process by healthcare professionals using online questionnaires to identify potential undesirable adverse effects and assess abortion success. If any problems arise at follow-up, patients are redirected to in-person services.²⁵ As such, this TM model can be used to support the woman throughout the entire medical abortion process, or parts of the process with few or no in-clinic visits required.

During the piloting of the intervention, Profamilia identified challenges and opportunities that informed the design of further implementation strategies. These strategies include the training of both Profamilia and external providers in remote areas to prescribe the kit if needed, hiring nurses to support the counselling service, refining and adapting Profamilia’s internal process for abortion care, and the creation of a new mass-media promotional campaign to increase diffusion of the telemedicine service among the Colombian population.²⁵

Outcomes and theory

To explore and illustrate the factors that affect the implementation of TM for medical abortion, two relevant implementation outcomes were chosen: feasibility and acceptability, defined according to Proctor et al.²⁸ They are often considered at initial implementation stages, as they influence the likelihood that a practice is adopted and successfully delivered in a certain setting.²⁸

The normalisation process theory (NPT) explains how new practices are operationalised, routinely embedded and sustained in a certain context.²⁹ NPT has been largely used to assess feasibility and it was chosen for its applications in qualitative research studies to describe and understand providers’ perceptions as they implement a new practice or intervention.³⁰ The constructs of NPT were used during framework analysis to inform the development of themes and categories from inductively derived codes.

Sample and participants

Key informants were employees of Profamilia at the time of data collection and were involved with TM services for medical abortion either as a healthcare provider delivering the service or as administrative staff coordinating the implementation of the service.

We used purposive maximum variation sampling followed by snowball sampling to identify and recruit a varied sample of key informants in terms of professional experiences, backgrounds, and roles in the delivery of TM services. The chosen sampling strategy aimed to generate a wide range of experiences and views on the implementation of TM for medical abortion among participants.

Data collection

Co-investigators from Profamilia facilitated recruitment by providing a list of individuals involved in TM abortion provision in the organisation, alongside their contact information. Co-investigators from Profamilia who facilitated recruitment were not involved nor had any influence on participants’ employment within the organisation and did not contact participants prior to or during data collection. To avoid ethical conflicts, recruitment and data collection were conducted by NP, a researcher who was external to the organisation. Participants were not, in any instance, pressured or forced to participate in the study by their employer.

Potential participants were informed about the study via email and invited to participate by NP. In case of no response, a follow-up email was sent to enquire about their interest in the study. All potential participants were provided with detailed written and verbal information about the study procedures, that participation was voluntary, and that they had the right to withdraw their participation at any time without any negative consequences. All participants provided verbal and written electronic consent before the interviews started.

A semi-structured interview guide (Annex I) with open-ended questions that explored participants’ views on access to safe abortion in Colombia, their experiences, and opinions of implementing TM for medical abortion, was used during interviews. The interview guide was pilot tested during the first two interviews with no changes required. Interviews were conducted online, via Zoom, in Spanish language by NP between March and October 2021. The interviews, which were conducted during participants’ working hours, lasted between 25 and 90 minutes and were audio-recorded. Data collection stopped when data saturation was achieved, that was when no additional data added new information to the findings.³¹

Analysis

Data were transcribed verbatim and pseudonymised. Interview transcripts were analysed in Spanish using the framework method.²¹ Codes were developed inductively from the data, and themes and categories were developed both inductively from participants’ accounts and deductively, following the constructs of the normalisation process theory.²⁹

Data were coded and analysed in five stages, as described by Gale et al¹⁹: first, familiarisation with the data through transcription and reading; then an analytical framework was developed by NP and ES with a set of codes and definitions. Researchers independently coded three interviews and met to discuss and build an initial coding framework. This framework was then used to independently code another three interviews, whose findings informed further revision and refinement of such coding framework. Finally, NP and ES indexed the remaining interviews according to the analytical framework and charted the data into a framework matrix. This matrix facilitated the interpretation of the data by NP. The interpretation process was influenced by the initial research questions and the constructs of the Normalisation Process Theory.²⁹ Analyses were conducted in Microsoft Word and Microsoft Excel and illustrative quotes identified during analyses were translated to English by NP.

Ethical considerations

This study adhered to the Helsinki Declarations for research ethics. Ethical permit was sought from the Comité de Ética en la Investigación de Profamilia (CEIP), [CEIP-04-2021] obtained on February 23rd, 2021, and from the Swedish Ethical Review Authority [2021-01154] obtained on March 24th, 2021.

Results

A total of 15 participants were included in the study. Of those, eight were recruited via purposive sampling and seven through snowball sampling. Participants worked in Profamilia clinics across three different cities, servicing urban, peri-urban and rural areas. Participants included five doctors, three psychologists, two coordinators, two nurses, two nurse assistants and one member of administrative staff, with varied lengths of experience in working with abortion, ranging from 6 months to 14 years. Background characteristics and demographic information are shown in Table 1.

Table 1:

Participant characteristics

Participant characteristics	N
Gender	15
Men	5
Women	10
Age	13
Age (mean, range)	40, 25-60
Clinic location	14
Bogotá	8
Medellín	4
Apartadó	2
Role	15
Coordinator	2
Doctor	5
Nurse	2
Nurse Assistant	2
Administrative Assistant	1
Psychologist	3
Years of experience with abortion services	14
Years of experience (mean, range)	4.3y, 3m-14y

The analysis generated three themes: (1) Understanding TM for medical abortion as a pragmatic solution for increased access in a conservative context; (2) We need to implement: willingness, needs and capabilities for TM for medical abortion, and (3) TM for medical abortion is a collaborative process of implementation, evaluation, and adaptation. Figure 1 illustrates how the themes contribute to an understanding of the theory constructs.

Figure 1:

Findings in relation to the constructs of the NPT, based on May and Finch²⁹

Understanding TM for medical abortion as a pragmatic solution for increased access in a conservative context

All participants ascribed meaning to the use of TM for medical abortion as an innovative and advantageous solution in the context of a global health emergency, the COVID-19 pandemic, and in a country where social, economic and geographical barriers to abortion care persist. Interviews revealed that the socio-political context had a strong influence on the feasibility of TM for medical abortion in Colombia and determined its acceptability among the participants. Participants construed TM abortion care as a desirable but somewhat uncertain alternative to in-person medical abortion. That is, most participants praised its advantages in terms of access, effectiveness, safety, comfort and comprehensiveness, although fear of complications and risk of misuse somewhat overshadowed acceptability of the practice, particularly among less experienced clinical providers.

As described by participants, TM for medical abortion was developed and implemented as a response to the COVID-19 pandemic and to increase access to safe abortion at a time when movement restrictions were widespread. In addition, competitive market dynamics in the provision of these services were reported to also have fostered implementation. According to participants, the pandemic facilitated contextual changes, including loosening of regulations, in an otherwise conservative context, where legislation around telemedicine and abortion is strict, and where abortion stigma and anti-abortion movements feed into and take advantage of a general lack of awareness around abortion to influence access to abortion services.

Participants relayed how taboos and stigma around abortion remain pervasive in the country. Anti-abortion movements stigmatise abortion services and harass abortion-seekers, thereby negatively impacting in-person access.

“Here in Colombia … we have some groups that are against provision of abortion services. They are named ‘pro-life’ but they are anti-rights (…) and forty days before Easter, every single day they stand here in front of Profamilia with signs saying, ‘no to abortion', ‘end abortion', ‘abortion is a crime', and I don’t know what (sic). And they stand with little dolls and everything, to create mental barriers to women who come to the clinic to receive the service.” (Key informant 1)

Thus, many participants see the use of TM for medical abortion as an acceptable alternative for those abortion-seekers who want to bypass stigma and preserve their privacy and confidentiality. However, the perceived power of anti-abortion movements also manifested in some participants’ fear of sabotage and backlash, as illustrated in the following quote:

“Here there are associations against abortion, mainly Christian and so on that, for example they can be registering and applying for the pills only to throw them away.” (Key informant 5)

Policy regulation on telemedicine service delivery were largely seen by participants as a limitation to the provision of flexible and person-centred services, particularly when conflated with strict regulations for medical abortion and the lack of political willingness by other health actors to facilitate access to abortion. These regulations often collided with the social, economic and geographic realities of abortion-seekers in most need of TM. For example, some participants mentioned that a requirement on the use of videocalls to deliver telemedicine, instead of regular phone calls, meant that the majority of abortion-seekers from low-resource, rural or marginalised communities, including adolescents and girls, were not able to access TM services due to their limited access to such technology and lack of quality Internet network.

“ … there were places where the signal [was] very bad. The connection failed many times (…) We had users telling us that they took [the consultation] from the highest top of a little mountain to be able to receive signal and receive the [medical] attention there.” (Key informant 3)

Legal requirements for the delivery of abortion medications, and limits on gestational age for medical abortion, were also described by several participants as negatively impacting TM services in Colombia. Given the time-sensitivity of medical abortion, rural and indigenous communities living in isolated and remote areas were particularly vulnerable to failures in delivery due to adverse weather events and episodes of social unrest.

“There were protests, there was a road blockage, difficulties in access for several days, and it was scary for the deliveries that we had, as they were delayed. So, we couldn’t arrive on time, when the user needed it.” (Key informant 12)

Some participants pointed out the influence that myths, expectations and lack of awareness about the abortion law had on access to abortion through TM. Further, they highlighted how limited knowledge surrounding the menstrual cycle and pregnancy in some communities and populations such as adolescents, could limit the usefulness of TM as many abortion-seekers would not be eligible for the service without being able to give details about their last menstrual period. There were also some concerns about the misuse of TM services as it would increase access to abortion pills, as described by one participant:

“When we talk about, for example, issues of violence, sometimes a man may be asking on behalf of a minor, or a woman on behalf of a minor, who are the perpetrators [of sexual violence]. They need to force her to have an abortion because it’s a danger for them. Sometimes the partner who is older than the woman and coerces her to have an abortion because when they had sexual relations it wasn’t consented … ” (Key informant 5)

Interviews revealed a diversity of opinions and perceptions about the benefits of TM for medical abortion, as compared to in-person abortion care, and it was not unusual for one person to hold a variety of sentiments about the service. While many considered TM for medical abortion as an innovative and advantageous practice with positive impacts on access, some still had reservations regarding its safety and therefore perceived it as dangerous or less desirable than in-person abortion care. Concerns about safety were more prominent among less experienced health care providers and grounded in the inherent nature of telemedicine where no in-person contact with the patient is made. For example, eligibility screening with ultrasounds and beta-hCG testing, and lack of in-person post-abortion follow-up were aspects which led some providers, particularly those with less clinical experience, to question the feasibility and appropriateness of this mode of service delivery.

We need to implement: willingness, needs and capabilities for TM for medical abortion

Interviews revealed heterogeneous accounts on organisational and individual readiness to implement TM for medical abortion. Participants drew on previous experiences of ad-hoc telemedicine work within Profamilia to describe the strengths and weaknesses of both the organisation and the providers of TM. They discussed persistent uncertainties around how to deliver the service, particularly in the first interviews, which took place in the first phase of implementation.

Previous successful telemedicine work and the strong organisational experience in providing SRH services influenced the development of the new and improved TM service for medical abortion. Participants appeared confident about the feasibility of TM for medical abortion in Colombia, given their previous experiences in providing the service and improvements made to the internal process for delivering abortion care:

“ … I would believe that this already, with all this … year of experience that we have got in the topic I wouldn’t think [there will be challenges]. I would say that as time passes, we learn from (…) things and processes we face on the way, so … And this helps to improve. So, I would think that, at this point, we could not waver in that aspect.” (Key informant 13)

Participants emphasised the human and material resources of the organisation to implement TM for medical abortion, and the actions undertaken to ensure providers’ readiness. For example, some participants confirmed the availability of trained providers and technology to deliver abortion through TM and the organisational capacity to absorb an expected increase in workload due to the implementation of the new service. In contrast, a few participants identified scarcity in human resources to be problematic, especially in relation to the initial contact with the abortion-seeker, while consultations with medical doctors seemed to be better coordinated. Participants constructively criticised existing gaps and areas for improvement in service implementation, as reflected in the quote below.

“ … we aren’t so effective. And it’s not because we don’t want to, I know my colleagues are phenomenal in what they do. But they can’t. They can’t because it is not possible to answer phone calls, manage WhatsApp, manage the chat. Because I am only one person doing this same process and it is very difficult. It is very difficult, and we are very few.” (Key informant 13)

Part of the relational work that the organisation undertook to ensure engagement in new interventions included building the capacity of staff involved in implementation. In this case, healthcare providers’ capacity in relation to abortion and use of technology was strengthened.

“ … for tele abortion I also received specific training before doing it.” (Key informant 10)

However, while some participants highlighted the responsibility of the organisation when it comes to disseminating new initiatives and providing training prior to implementation, others complained about the uncertainty and lack of clarity about their roles in the intervention, its different steps, and its specificities for certain population groups. The need for structure and protocol was highlighted by a young doctor:

“ … It is very important to have everything very well structured and that all of us who deliver tele abortion speak the same language (…) that we all are under a protocol, that everything is strictly followed so that it works, increases access to health services and is safe.” (Key informant 10)

Some participants brought up the need for protocols and guidelines and insisted on the uncertainty around implementing TM due to the lack of guidance in the country and international referents, an opportunity taken by Profamilia to lead on the implementation of TM for medical abortion in Colombia.

“I told my boss ‘Doctor, one must sit here with the uncertainty on how to do something where there's nothing written and where no one has done anything'.” (Key informant 1)

Interviews highlighted the importance of clinical experience to providers’ confidence and willingness to deliver medical abortion care via TM. Some participants, particularly those with longer experience and trust in their own and others’ ability to deliver services through TM, held positive views and high expectations on its implementation, while others, less experienced and less familiar with abortion through TM, remained cautious.

Participants also identified the supportive and innovative leadership within the organisation as a key factor that determines the feasibility of TM for medical abortion, as a key informant exemplified:

“ … I think that the commitment of the organisation and the top management has been fundamental for this project.” (Key informant 2)

TM for medical abortion is a collaborative process of implementation, evaluation and adaptation

Interviews revealed that the implementation of TM for medical abortion was a collective and collaborative process that is constantly evolving through feedback loops of evaluation and adaptation. Participants referred to continuous collaboration, both internally within the organisation as well as externally with other entities and providers, as a factor that determines the feasibility of the telemedicine service.

Teamwork was reported to be a core organisational value that contributed to the feasibility and successful implementation of a telemedicine service that requires an interdisciplinary and coordinated approach:

“ … It is a common denominator, teamwork, this is one of our organisational values (…) and I think this has been one of the tools that allowed to carry out this collaborative work.” (Key informant 2)

Some participants, those in coordinating roles in particular, pointed out the importance of external relationships with a variety of entities to increase feasibility of implementing TM by expanding its reach, availability and accessibility to abortion-seekers. These collaborations were diverse in nature and included commercial agreements with delivery companies, financial agreements with health insurance entities, and strategic and technical alliances with other non-profit organisations and donors.

Some participants described how implementation steps were constantly monitored and collectively appraised to identify complications and respond to them. Thus, as implementation progressed, participants reported that adaptations to service delivery protocols and steps of the TM process were introduced both by individual providers and by the organisation itself. These changes and adaptations to the operationalisation of the service were inextricably linked with formal and informal evaluation mechanisms and were based on providers’ experiences and TM users’ behaviours and feedback.

“So, [to avoid TM users from taking unnecessary follow-up tests] we had to adjust our protocol too. We recommended users not to take pregnancy tests after 15 days, but after 18 or 20 days, so that the pregnancy test is truly negative and further follow-up processes are not needed.” (Key informant 12)

As benefits became apparent during implementation and safety and effectiveness of medical abortion through TM were confirmed by their own experiences, most participants reported being satisfied with the new service. However, some participants referenced significant logistical challenges, issues of safety and limited functionality of TM as experienced during implementation. They also suggested further adaptations to adjust and strengthen certain procedures such as follow-up and communication with TM users.

Discussion

Our findings show that implementation of TM for medical abortion is a dynamic and non-linear process that is favoured by strong leadership, collaboration between healthcare providers with an appropriate skill mix and varying lengths of clinical experience, and by organisational investments in training, appraisal, and adaptation. In the early phase of implementation, the use of TM for medical abortion was found to be highly acceptable, although some concerns existed surrounding medical safety, effectiveness, and a perceived risk that the services do not have appropriate safeguarding mechanisms in place. We identified social, political, and economic factors such as anti-abortion movements and contradictory regulation on mode of communication required for TM service delivery, that impede implementation efforts and stymie the potential of this intervention in reaching those in need.

Findings from this study underscore the need for collaboration and teamwork to ensure the successful implementation of the TM service. Two recent studies in the US reported similar results – internal and external collaboration were found to be critical to successful implementation of TM for medical abortion in different modalities.^32,33 An implementation study of a site-to-site telehealth abortion service that resembles initial pilots and ad-hoc TM work by Profamilia, found that teamwork between remote clinicians and health centre staff was strengthened as a result of implementation.³² This aligns with participants’ accounts of teamwork as a necessary approach to provide TM services. Our findings also echoed those of Godfrey et al.,³³ as collaboration with external organisations for informational, logistic and funding support for TM services for medical abortion was a factor that contributed to implementation success in both studies.

Despite growing evidence that TM for medical abortion is safe and effective, some participants questioned the medical safety of TM in the absence of in-person eligibility screening and follow-up consultations.^5,6,34 Scholars have suggested that concerns about medical safety in relation to TM for medical abortion may stem from a perceived loss of control over a traditionally medicalised process.³⁵ We found that willingness of healthcare providers and administrative staff to deliver abortion services using TM, combined with extensive clinical experience, favoured implementation. Similar results were reported by Godfrey et al. in the US, who claimed that provider and staff self-efficacy, motivation, passion and determination to deliver abortion care via TM were success factors for implementation.³³ The influence of experience on providers’ acceptability has been emphasised in previous studies, in which experience in delivering abortion care using TM led to learning and increased confidence in the service.^36,37 Based on our findings and previous research in the US and Scotland,^33,36 it is clear that capacity building and senior support play a key role in the implementation of TM services. This highlights the importance of ensuring a sufficient skill mix and involving healthcare providers with different durations of experience, in future endeavours to implement abortion service delivery using TM. Recently published guidelines by WHO² and the Royal College of Obstetricians and Gynaecologists,³⁸ featuring a recommendation around the use of TM for medical abortion and best practice guidance for TM service delivery, may ease healthcare providers’ concerns. It is worth noting that, following publication of WHO guidelines, Colombia’s Ministry of Health updated their guidelines to allow medical abortion until 12 weeks of gestation.³⁹ Still, our findings underscore how early implementation of TM for medical abortion must emphasise issues of safety and effectiveness to bring everyone on board.

Our findings underline the perceived lack of safeguarding mechanisms in TM services for abortion-seekers experiencing reproductive coercion and intimate partner violence (IPV). Concerns and arguments around safeguarding and TM are part of a current debate often focused on abortion coercion, that is sometimes used as an excuse to end or limit abortion access through telemedicine.⁴⁰ Although evidence shows that these concerns are easily debunked, it is evident from our findings that these are real concerns among implementers of TM for medical abortion, impacting acceptability of the service delivery modality.^40,41 When TM was introduced in the UK, safeguarding referrals increased, indicating that with experience and appropriate training, healthcare providers can remotely identify potential signs of abuse and coercion.^36,40 Efforts to train implementors of TM for medical abortion on how to ensure safety of TM users and identify those in need of IPV services, must therefore be part and parcel of pre-implementation activities and in-service training. This is especially true in settings where the prevalence of IPV is high, such as in the Latin American region.⁴²

Our study results suggest that social and institutional opposition to abortion may have negative consequences for TM services in Colombia, as has been reported in the UK.⁴⁰ In Colombia, politicians, anti-abortion groups and associations of physicians who are conscientious objectors, have expressed their concerns about deregulation of the use of TM for medical abortion during the pandemic, both in the media and in the Colombian Congress.⁴³ Widespread implementation of TM will increase the proportion of safe and legal abortions in the country which means a hike in reported rates of abortions. Feminist groups and NGOs in Colombia fear a backlash by anti-abortion groups who will use this to push for restrictions on the use of TM and a rollback of reproductive rights.⁴³ However, it is the advocacy by these feminist groups that has allowed the recent legislative changes in the country that facilitate access to abortion until week 24 and that create an enabling environment that removes barriers to service provision.^39,44,45 Forthcoming publications and reports about TM for medical abortion in Colombia, supporting its use, may help cement this intervention as a permanent feature of available abortion service delivery modalities in the country, in the private as well as public health sector.

Quality abortion care services require an enabling environment, which includes a supportive framework of law and policy.² Our interviews revealed a regulatory policy environment that limits access to TM abortion services, particularly among marginalised populations and low socio-economic communities. Similar findings were reported by Ortiz et al.,⁴⁶ who found that regulatory authorities in Colombia required electronic signatures or audio-recording of the abortion-seeker’s verbal consent. Concurrently, the Colombian Ministry of Health states that different sorts of information and communication technologies can be used to deliver telemedicine and electronically signed consent is not required.⁴⁷ This highlights a significant inconsistency between policy and practice, which disproportionately impacts access among those who need it most. This policy-practice gap and enforcement of a regulation that is contrary to the government's own policy constitute unnecessary barriers for provision of and access to abortion care using TM. Of note is that the Ministry of Health’s Motion 2654/2019 explicitly stated that TM for medical abortion aims to address inequities in access for the nations’ marginalised populations.⁴⁷ It appears there is an urgent need to clarify the regulatory framework and close the policy-practice gap, ensuring that policies and their implementation are supportive rather than obstructive. Indeed, the Ministry of Health took some steps in this direction, by placing a duty on service providers to ensure availability of TM services.³⁹

Innovative digital solutions such as TM, can be important tools in advancing sexual and reproductive health and rights. To optimise the implementation of such interventions, lessons learnt from early stages of implementation may be useful to understand the needs for further adaptations and how to address implementation bottlenecks. Our findings revealed factors that may limit and factors that may facilitate implementation and scale-up of the use of TM for medical abortion in the Colombian context, and highlighted strengths on which decision-makers can capitalise to target adaptation efforts and further implementation strategies. Further studies are needed to understand how issues surrounding acceptability and feasibility are impacted over time by studying how views and perceptions change during the implementation process. In addition, qualitative research is needed to understand users’ views on TM for medical abortion in Colombia as data is scarce in this regard.

Strengths and limitations

Our findings should be understood within the limitations of the study. Data collection mainly took place at the initial stages of implementation. This limits our understanding of the effects of long-term implementation of TM on feasibility and acceptability once the practice is normalised and integrated into routine care. Furthermore, all respondents had agreed to provide or, in some cases, been part of initiating the TM service, which means that these respondents may have had a more favourable view on TM for medical abortion from the start, impacting on our findings relating to acceptability.

The main strength of this study relies on the rigour and transparency of the framework method and the application of theory during analysis and interpretation, which strengthens the trustworthiness of the findings and make the findings transferable to other similar settings. Since the study was conducted within a specific organisation in Colombia, unique for this setting, other organisations attempting to implement TM for medical abortion in the Latin American region may not share the same implementation experiences. Nevertheless, our findings provide important learnings from the early stages of the implementation process that others may make use of when implementing or scaling up TM for medical abortion.

Another strength of this study is the sampling strategy which enabled a varied sample in terms of professional experiences and backgrounds, ranging from coordinators to administrative personnel and healthcare providers. This wide range of participants allowed us to look at the implementation of the innovation from different angles and contributed to triangulation of the data.

Conclusion

Our findings indicate that at an early stage of implementation, TM for medical abortion was positively viewed and deemed feasible by healthcare providers and others involved in its implementation in Colombia. Pre-implementation training and clinical experience among those implementing and coordinating the service, as well as a strong leadership, played a central role in overcoming implementation bottlenecks at the individual and organisational level. In Colombia, TM for medical abortion can overcome barriers in access to abortion care. However, concerns for medical safety among implementers and contextual challenges, such as the regulatory environment, communication-related limitations and social opposition to abortion, remain to be addressed to allow TM for medical abortion to live up to its full potential. Lessons learnt from this study may be useful for those wishing to implement or scale-up TM for medical abortion with the goal of realising sexual and reproductive health and rights for all.

Appendix.

Annex I.

Interview guide

Subject	Questions
Introductory question	How do you think the Covid-19 pandemic has impacted on girls and women’s access to safe abortion care in Colombia? Problems faced by abortion seekers. Reasons. Interventions or mechanisms put in place for abortion access. Opinions. Experiences with patients during the Covid-19 pandemic
Feasibility	Could you tell me about any challenges you may have encountered since you started delivering abortion through telemedicine and how have you addressed them? During pre-abortion counselling? When prescribing the drugs? When sending medication? At post-abortion follow-up? Could you tell me about a (typical) situation/case in which abortion through telemedicine worked particularly well? Why? Could you tell me about a (typical) situation/case in which abortion through telemedicine was particularly hard to deliver? Why?
Acceptability	Could you tell me about what it is like to work with abortion through telemedicine? Do you have any concerns? What are they? What are your personal thoughts and opinions about abortion through telemedicine? Do you think it is working well? Why? Why not? What impact do you think it has on access to safe abortion?
Acceptability and feasibility	How would you describe the quality of the service you provide via telemedicine as compared to on-site abortion services?
Closing questions	Would you like to add anything, or do you feel that there is anything important that I have not brought up? We have now reached the end of our interview and we are thankful for your responses. Do you have any questions or final comments?
Sociodemographic information	Age Gender Profession or role (doctor, nurse, coordinator, etc.) Where do you work? Name of city and whether the area is urban/rural. What type of population do you work with? (in terms of income) How long/since when have you been working with abortion? How long/since when have you been working with abortion through telemedicine?

Funding Statement

This study was funded by the Department of Women's and Children's Health, Karolinska Institutet, Sweden.

Author contributions

NPF, MCJ, GG and AC conceptualised the study. NPF and MCJ recruited participants, NPF conducted interviews. NPF and ES analysed findings and drafted original manuscript with support from AC. MCJ, GG, ASR, PM, and DCM provided resources and validated findings. All authors provided input on the manuscript and approved the final draft. The project was supervised by AC.

Disclosure statement

No potential conflict of interest was reported by the author(s).