How we do it


The study is a 3 country (Ethiopia, Malawi, Zambia) comparative study design of contraception and abortion services for adolescents. Within each country, two contrasting study sites (tertiary hospital and designated adolescent sexual and reproductive health services [ASRHS]), have been identified. The study design permits three analytic levels: within-country by facility; cross-country; and cross-facility. The 3 countries represent a range of ASRHS and abortion legal frameworks and permit a maximum difference design. It will focus on evidence from adolescents who are either seeking post-abortion care from an abortion initiated elsewhere (at home, unregulated sector, etc.) or who are seeking an abortion. In doing so, we will be able to understand non-facility-based care-seeking behaviours. By comparing evidence generated from two different facilities in each country (standard tertiary hospital care vs. ASRHS), our study design will permit a more detailed understanding of the implementation factors (including barriers) that lead adolescents to seek care from different providers in the same setting. We will use 4 research methods to generate multiple perspectives to develop the most rounded understanding of how to improve the implementation of safe abortion services for adolescents, with multidisciplinary analyses to maximise research uptake:
1. Comparative policy, health system costing and legal analyses: To address RQ1, RQ3c. Country-level context and other factors influence implementation of ASRHS, including: social, cultural, economy, political, legal and health systems environments. We will conduct a comparative cross-country analysis of contextual factors in order to draw out the potential for cross-national learning within the team.
Analyses: Comparative cross-country thematic analyses of secondary data, including: policies, protocols, laws, and evaluations. These comparative analyses will use the policy triangle (context, content, actors and processes) framework in order to develop retrospective and prospective profiles. These profiles will be updated over the course of the research project to incorporate emergent context changes, and will be used to inform the content of the key informant interviews (Method 4 below) and our research uptake. Understanding the costs associated with unsafe and safe abortions requires multiple perspectives and approaches. Building on approaches and evidence generated by the team we will model and develop costing measures, including: i) the health system costs of preventing unintended pregnancies with contraception versus the cost of treating complications of unsafe abortions or providing safe and legal abortions to adolescents in need of these services in each country ii) The estimated costs of scaling up intervention approaches recognised by adolescents as being most acceptable and important to them in each country. iii) The financial costs of safe and unsafe abortion care-seeking and receiving for adolescents in each country. iv) The social costs of an unwanted pregnancy and childbirth for adolescents in each country.

2. Quantitative survey of adolescents: To address RQ1, RQ2. To establish which aspects of implementation act as a barrier or facilitator to adolescents’ use of contraception and SA/PAC services. Facility-based recruitment of adolescents seeking either SA or PAC following an abortion initiated elsewhere. In each country, adolescents (n≈2883) seeking care at two public sector facilities (tertiary hospital vs. ASRHS) will be recruited. We focus on the public sector because it is where most vulnerable or marginalised adolescents seek care. Over a fixed time period in each country, varying according to expected numbers of adolescents seeking care (Ethiopia-4 months, Malawi-8 months, Zambia-6 months), all adolescents identified as having sought either SA or PAC by a study-trained senior nurse will be invited to participate once ready for discharge. We will include adolescents who received treatment as out-patients and those hospitalised for severe complications. Interviewers will work across a range of shifts (Mon-Fri vs weekend; day vs. night) to maximise the range of adolescents seeking care. Research instruments already developed by the team will be adapted and tested with attention to inter-country comparability. The research instrument will generate evidence on: detailed care seeking pathways (and their influences and influencers); barriers to care-seeking (eg: knowledge, confidentiality, cost, transport, unofficial provider payments, perceived quality of care); sociodemographic status; contraceptive (non-)use; direct service costs (for example, fees per procedure or intervention); indirect costs (e.g.: travel, food, lost productivity); resources used to pay costs (e.g.: credit, asset sale, borrowing, loss of wages); knowledge of the law, including understanding of adolescent rights to services; barriers and facilitators to care-seeking. Interviews will be conducted in a private office by female research assistants (RA) fluent in all major local languages in each country. RAs will be been trained in research ethics, informed consent and interviewing techniques. An established two-interviewer approach will be used: one RA will conduct the interview in a conversational style to put the participant at ease and facilitate the narrative flow, whilst a second RA will complete the research instrument. Towards the end of the interview the second RA will ask supplementary questions not covered by the first RA to ensure completeness. Treatment records will be accessed, with permission, to validate individual reports of abortion care received and morbidity symptoms as a result of unsafe abortion procedures or attempts. This is an approach that has been used successfully by team members to generate high quality evidence about pathways to care-seeking.
Analyses: Statistical analyses by and within country, by and within treatment route (SA vs. PAC), by and within facility. Trajectories to abortion care are a distinct group of healthcare-seeking behaviours because issues of legality and understanding of legal rights overlay an individual’s pathway to care. Analysis of the steps from the decision to terminate a pregnancy to securing abortion care provides insight into the implementation of services from the perspective of users’ care-seeking behaviour and influencing factors, including barriers and delays to accessing care, stigma, and perceptions of risk and care.

3. In-depth qualitative interviews with adolescents: Recorded (with consent) interviews with a sub-sample of adolescents drawn from the quantitative survey (n=30 per study site). Selection will maximise heterogeneity in trajectories to accessing abortion care to establish the range of reasons why adolescents did (not) use safe abortion services, their sources of information, delays to care-receiving and pathways into the health system. These interviews will focus in detail on barriers to care-seeking, and are likely to include topics such as: knowledge about where and when what services are available; issues related to confidentiality, privacy and anonymity; cost, including transport and unofficial provider payments; ability to care seek (eg: time off school); and perceived quality of care. We expect to conduct about 30 interviews per facility; the final sample size will be determined by reaching a sufficient analytical understanding.
Analyses: Verbatim translation and transcription of recorded interviews will be analysed using framework analysis, to systematically categorise, organise and synthesise case-oriented qualitative data. Preliminary content analysis of a sub-set of purposively selected transcripts will be discussed in a team analysis workshop. The set will maximise heterogeneity, including: age, marital status, ethnicity, education, employment, residence, wealth, clinical intervention and outcomes. Findings will inform the design of the framework analysis. Key recurring themes and related sub-themes will be identified and mapped to ensure conceptual clarity within them, with no obvious omissions or overlaps. The themes will be shaped by the analytical ideas that emerge from the preliminary whole team content analyses, meaning that a combination of deductive and inductive themes will be included.

4: In-depth repeat interviews with key informants: Key informants (n≈20 per country) involved in the implementation of contraception and abortion services for adolescents, purposively sampled for a range of perspectives (nurse-midwives, doctors, administrators, pharmacists decision-makers, funders, donors, civil society [including faith-based organisations]), and reflect the comparative (standard vs. ASRHS) design of interviews with adolescents. Key informant interviews (KIIs) will focus on a range of topics, including: funding for SRH in general and ASRHS in particular; training in ASRH for health professionals; levels of awareness of ASRH policies; capacity in planning, implementing and monitoring ASRH activities at national and sub-national level; objection to abortion service provision (in general, and for adolescents in particular); and, attitudes towards ASRHS. Eight implementation outcome variables will be used to organise these interviews in order to provide insights into how implementation and barriers can be improved [1]: i) Acceptability (perception among stakeholders than an intervention is agreeable); ii) Adoption (intention / initial decision / action to employ an intervention (or scale it up)); iii) Appropriateness (perceived relevance of the intervention in a setting); iv) Feasibility (the extent to which an intervention can be carried out in a particular context); v) Fidelity (the degree to which an intervention was implemented as it was designed in an original protocol/plan/policy); vi) Implementation cost (incremental cost of the implementation); vii) Coverage (degree to which the population that is eligible to benefit from an intervention actually receives it) and viii) Sustainability (the extent to which an intervention is maintained in a given context). The project will develop, test and apply a question guide for KIIs across the 3 countries using these topics. Key informants will be interviewed twice. Interview 1, prior to the facility-based survey, will reveal tacit knowledge and assumptions about the implementation of adolescent abortion services. Interview 2 after preliminary analyses of data generated from adolescents in order to improve the quality of research interpretation and findings, and the likelihood of research uptake. Recorded in-depth interviews (with consent) will be transcribed.
Analyses: Thematic analyses of verbatim transcripts of interviews. The transcripts will be analysed using a framework structured by the 8 implementation outcome variables.


Our mixed methods design will involve the collection and analyses of both qualitative and quantitative data on the implementation of ASRHS and their integration. Project time has been allocated to ensure that this integration is meaningful and produces a whole that is greater than the sum of the individual qualitative and quantitative parts. This will be an iterative process across the project, for example: comparative policy and legal analyses will contribute to the content of key informant interviews; analyses of adolescent interviews will inform repeat key informant interviews; reactions and responses from research uptake, including from the project Youth Advisors, will contribute to analyses of evidence from adolescents and key informants.

We will produce evidence to inform the provision of contraception and abortion services for adolescents in African countries. We aim to contribute to the evidence base for more realistic and improved provision, and ultimately a reduction in the deaths and morbidity attributed to unsafe abortion among adolescents. We will achieve this by focusing research uptake activities to as wide a policy, practitioner, academic and civil society audience as possible.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s