Table 2

Research gaps for self-administration of injectable contraception

Clinical, programmatic, values-based or human rights question		Current WHO recommendation	Strength of recommendation	Source or quality of evidence	Key GER considerations	Ideal study design(s)	Feasibility and practical constraints	Alternative study design	Methodological issues arising in the alternative study design
Benefits versus harms	PICO: should self-administered injectable contraception be made available as an additional approach to deliver injectable contraception?	Self-administered injectable contraception should be made available as an additional approach to deliver injectable contraception.	Strong.	Moderate.	No data available from LMIC; availability, accessibility, privacy and confidentiality.	RCT conducted in LMIC with inclusion of outcomes to measure accessibility, privacy and confidentiality.	As a strong WHO recommendation exists, it would be unethical to repeat an RCT of the intervention.	Prospective non-controlled cohort study to be conducted in several low- and middle-income countries where the intervention is or will be implemented. Analysis to combine quantitative survey supplemented with focused in-depth qualitative interviews with women (both those who have used and have not used self-administered injectable contraception and including adolescent girls and young women) to explore GER outcomes of availability, accessibility, privacy and confidentiality. Cost analyses can be nested in the cohort study.	Confounding: participants in cohort self-select to self-administer injectable and may have other similar characteristics that may influence their ability to access injectable contraception. Measurement bias: self-reporting bias an issue for qualitative component. This will require well trained researchers and validated tools. The triangulation of the qualitative data with the quantitative survey results provides an element of additional validation of outcomes.
Values and preferences	Is there an impact of differences between groups of women (eg, age, socioeconomic and occupational status) on their values and preferences with respect to self-administration of injectable contraception?	Not applicable.	Not applicable.	GDG.	Accessibility, privacy and confidentiality, non-discrimination.	Cross-sectional study of women using injectable contraception to evaluate associations between characteristics and use, or willingness to use.	A survey may not provide an objective measure or a true reflection of behaviours if driven by questions requiring socially acceptable responses.	This is a feasible study design but could be supplemented with diary collection of data as greater privacy and less likely for social desirability to influence self-recording. Partnering with women who have used or are willing to self-administer injectable contraception is important to identify critical and non-judgmental questions.	Measurement bias: for both survey and diary collection, questions will need to be formulated together with women using injectable contraception to ensure trust, participation and validity to avoid social desirability bias.
	What happens after women discontinue use of self-injectable contraception – do they use other methods?	Not applicable.	Not applicable.	GDG.	Accessibility.	National pharmacy or clinic-based registry of women using self-administered injectable contraception.	Registries can be costly to establish, maintain and monitor. A registry requires political will and capacity to provide effective data to inform decisions.	Convenience sample and survey of women attending family planning clinics to elucidate their prior modes of contraception and reasons for discontinuation.	Selection bias: such a sample would only include women who attend the clinic to change to a different form of contraception; women who cease to use self-administered contraception and do not return to the clinic would not be included. Their choices may differ from those women who do attend the clinic.
	What is the impact of stigma on the choice of self-administration of injectable contraception?	Not applicable.	Not applicable.	GDG and survey.	Accessibility, privacy and confidentiality and non-discrimination.	Qualitative study of interviews of women self-administering injectable contraception to elucidate their experiences and observations.	This is a feasible study.	Not applicable.	Selection bias: as with all qualitative studies, participants are purposively selected. However, this is an advantage in this study design as those women who have considered this issue will have rich experiences and observations to share.
	What are the optimal models of information provision for awareness raising and increasing knowledge of self-administration of injectable contraception?	Not applicable.	Not applicable.	GDG and survey.	Accessibility, the right to seek, receive and impart information.	Comparative effectiveness research of different models of information provision, optimally in an RCT.	This will first require development of models of dedicated information provision and health literacy (eg, telephonic, internet, posters and counselling) prior to experimental testing.	Exploratory qualitative interviews with women self-administering or willing to self-administer to identify their knowledge, attitudes and understanding of injectable contraception are required to inform models of information provision.	Selection bias: as with all qualitative studies, participants are purposively selected. However, this is an advantage in this study design as those women who have considered this issue will have rich experiences and observations to share which can guide the development of health promotion intervention and tools.
Human rights and equity	What implementation measures can ensure that inequity is reduced or minimised when self-administration is introduced?	Not applicable.	Not applicable.	GDG.	Non-discrimination.	Comparative effectiveness research of different models of implementation, optimally in a cluster RCT.	An RCT of models of implementation may be very costly due to the complex nature of implementation strategies that are dependent on the setting, type of provider and training required, and ensuring fidelity to the intervention.	Descriptive case studies of demonstration projects using different strategies to implement injectable contraception in programmes in several countries to indicate optimal measures to ensure equitable access and to formulate lessons learnt for scale-up elsewhere	Lack of generalisability: current practices and potential for scale will differ markedly between countries. Therefore selection of country sites for demonstration projects will require characterisation of underlying resource needs, training, staff and sociopolitical factors to ensure maximal diversity and that reasonable comparisons can be made between countries. Triangulation between data sources and sites will be necessary to increase generalisability.
Acceptability	Is there an impact of differences between healthcare providers (eg, age, income status of country, private/public sector) on the acceptability of self-administration of injectable contraception?	Not applicable.	Not applicable.	GDG.	Acceptability and accessibilty.	Cross-sectional study of healthcare providers to evaluate associations between characteristics and willingness to prescribe and provide injectable contraception for self-administration.	A survey may not provide an objective measure or a true reflection of behaviours if driven by questions requiring socially acceptable responses.	This is a feasible study design but could be supplemented with focus groups or qualitative interviews.	Measurement bias: for both survey and interviews, questions will need to be formulated together with healthcare providers to ensure trust, participation and validity and to avoid social desirability bias.
Acceptability	What is the scale and consequence of incorrect use of self-administration?	Not applicable.	Not applicable.	GDG.	Accountability.	National pharmacy or clinic-based registry of women using self-administered injectable contraception.	Registries can be costly to establish, maintain and monitor. A registry requires political will and capacity to provide effective data to inform decisions.	Sentinel (active) surveillance through monitoring and evaluation of adverse effects and adverse events at clinic or provider level embedded in local and national quality assurance programmes.	Reporting bias: the data collected is dependent on the training and expertise of providers to adequately recognise adverse events. Resources can be directed towards appropriate sentinel sites to ensure staff are skilled and that data integrity is ensured.
Resource use	What are the user and provider costs of self-administration (as well as out-of-pocket expenditures), compared with provider-based administration?	Not applicable.	Not applicable.	GDG and survey.	Accessibility.	Empirical costing studies to estimate user and provider costs, and financing sources for each (users pay or health system pays) Data are currently only available from one study in three countries in sub-Saharan Africa. Given that costs vary by context and by delivery model, there is a need for more representative cost estimates to confirm that user costs and out-of-pocket expenditures are likely to decrease with self-administration.	This is a feasible study but can be costly to conduct across a representative sample of users and facilities.	This is feasible but could be supplemented by modelling based on the existing empirical cost data, cost structures and expert consultation on expected variation in parameters.	Measurement bias: the results are dependent on user recall and self-reported resource use, opportunity costs, as well as on assumptions regarding shared costs in health facilities.
	What is the cost-effectiveness of self-administration, compared with provider-based administration?	Not applicable.	Not applicable.	GDG.	Accessibility.	Modelling studies of cost- effectiveness across broader range of settings, taking both a healthcare provider and societal perspective.	This is a feasible study but its quality and usefulness depends on the reliability of the parameters for unit costs and effectiveness, which would be drawn from costing studies (noted above) and RCTs. It will also require more long-term modelling to capture downstream effects on the types of contraceptives self-injecting women switch to compared with provider-based injecting women.		Measurement bias: the results are dependent on selection of variables included in the model, assumptions regarding discount rates, cost-effectiveness thresholds and accurate estimates of current and projected costs and effects.
	What is the environmental impact of disposal of self-administered injectable oral contraception?	Not applicable.	Not applicable.	GDG.	Accountability.	Collection and collation of reports from environmental regulatory bodies documenting scale and impact of medical waste and household waste.	Regulatory agencies are unlikely to differentiate between whether injectable contraceptives present in waste are sourced directly from clinics or from the homes of users.	A qualitative household diary study of women who self-inject in which they record waste from the packaging and their management of it. This could be delivered with the injectable packaging and women can return it when collecting their next prescription.	Recall bias: women may not remember what they did with the packaging if they do not record it at the time. This may require an incentive during the research period in order to encourage women to complete the household waste diary.

GDG, Guideline Development Group; GER, gender, equity and human rights; RCTs, randomised controlled trials.