Exploring Somali women’s reproductive health knowledge and experiences: results from focus group discussions in Mogadishu

Abstract

With a total fertility ratio of 6.7 children per woman, a maternal mortality ratio over 1,000 deaths per 100,000 live births, high rates of sexual and gender-based violence, and the lowest contraceptive prevalence rate in the world, women’s reproductive health indices in Somalia prove alarming. The voices of women living in Somalia have long been neglected and we undertook this qualitative study to explore women’s reproductive health knowledge and experiences. In 2014, we conducted four focus group discussions with 21 married and unmarried women of reproductive age living in Mogadishu, Somalia. Discussions took place in Somali and we used a constant comparative approach to analyse the discussions for content and themes. Our findings reveal that misinformation, restrictive policies, mistrust of clinicians, and prohibitively expensive services shape women’s experiences and health-seeking behaviours. Women identified the need for culturally resonant reproductive health information and services as a significant priority. As Somalia begins to emerge from over two decades of civil war, it is imperative that comprehensive reproductive health issues are included on the national agenda and that women’s perspectives are incorporated into future policies and interventions.

Résumé

Avec un taux de fécondité total de 6,7 enfants par femme, un taux de mortalité maternelle supérieur à 1000 décès pour 100 000 naissances vivantes, des proportions élevées de violence sexuelle et sexiste et le plus faible taux d’utilisation de contraceptifs au monde, les indicateurs de la santé génésique des femmes en Somalie sont préoccupants. Nous avons entrepris cette analyse qualitative pour étudier les connaissances et les expériences des femmes en matière de santé génésique. En 2014, nous avons organisé quatre discussions de groupe avec 21 femmes mariées et célibataires en âge de procréer qui vivaient à Mogadiscio (Somalie). Les discussions se sont déroulées en somali et nous avons utilisé une approche comparative constante pour analyser le contenu et les thèmes des discussions. Nos conclusions ont révélé que la désinformation, les politiques restrictives, la méfiance à l’égard des cliniciens et le coût prohibitif des services façonnent les expériences des femmes et leur comportement de recherche de soins. Les femmes ont jugé que le besoin d’informations et de services de santé génésique culturellement adaptés était une priorité. Les voix des femmes vivant en Somalie sont depuis longtemps négligées. Alors que la Somalie commence à émerger de plus de vingt années de guerre civile, il est impératif d’inclure des questions globales de santé génésique à l’ordre du jour national et d’intégrer les perspectives des femmes dans les futures politiques et interventions.

Resumen

Con una tasa de fertilidad total de 6.7 hijos por cada mujer, una razón de mortalidad materna de más de 1000 muertes por cada 100,000 nacidos vivos, altas tasas de violencia sexual y violencia de género, y la menor tasa de prevalencia de uso de anticonceptivos del mundo, los índices de salud reproductiva de las mujeres en Somalia son alarmantes. Desde hace mucho tiempo, se hace caso omiso a las voces de mujeres que viven en Somalia. Llevamos a cabo este estudio cualitativo para examinar los conocimientos y experiencias de salud reproductiva de las mujeres. En 2014, realizamos cuatro discusiones en grupos focales con 21 mujeres casadas y solteras en edad reproductiva, que vivían en Mogadishu, en Somalia. Las discusiones se llevaron a cabo en somalí y utilizamos un enfoque comparativo constante para analizar el contenido y las temáticas de las discusiones. Nuestros hallazgos revelan que la información incorrecta, políticas restrictivas, desconfianza en médicos y servicios prohibitivamente caros afectan las experiencias de las mujeres y sus comportamientos con relación a la búsqueda de servicios de salud. Las mujeres identificaron como una prioridad significativa la necesidad de recibir información y servicios de salud reproductiva culturalmente resonantes. A medida que Somalia empieza a emerger de más de dos décadas de guerra civil, es imperativo incluir en la agenda nacional asuntos relacionados con la salud reproductiva integral e incorporar las perspectivas de las mujeres en futuras políticas e intervenciones.

Keywords:

• reproductive health
• post-conflict settings
• contraception
• qualitative research

Introduction

Somalia’s civil war erupted in 1991, essentially devastating the already fragile public health sector and leaving up to 80% of the population without access to basic health services.^1,2 Somalia’s reproductive health indices reflect this overarching dynamic. The World Health Organization (WHO) and UNICEF estimate that about 1% of women in Somalia currently use a modern method of contraception.^3,4 Consequently, Somalia’s total fertility rate is estimated to be one of the highest in the world, at approximately 6.7 children per woman.^4,5 Only 9.4% of births in this context occur in health facilities,⁶ and the maternal mortality ratio in Somalia is over 1,000 deaths per 100,000 live births.^7,8 Access to a range of existing reproductive health services, such as ongoing contraceptive methods, antenatal care, skilled birth attendants, and postpartum care, is extremely limited.² Abortion is severely restricted in Somalia and is only legally permissible to save the woman’s life. And Somalia remains one of the only countries in the world yet to register a dedicated progestin-only emergency contraceptive pill.⁹ Compared to other countries in the East African region, Somalia remains an outlier with respect to reproductive health indices and outcomes; use of skilled birth attendants during childbirth and modern contraceptive methods, for example, is considerably higher in every country throughout the region.^10,11

Rates of sexual and gender-based violence are critically heightened in conflict and post-conflict settings.¹² Rape is often used as a weapon of war, particularly among the most vulnerable of women in these contexts, and has become a pervasive problem throughout much of Somalia.^2,13 Rape often goes unreported and research indicates that women are generally unaware of the types of services that can be sought to prevent pregnancy and sexually transmitted infections.² Displacement and forced migration has contributed a great deal to the dynamics surrounding sexual and gender-based violence in Somalia. Approximately one tenth of Somalia’s population is internally displaced⁴ and it is estimated that Mogadishu alone houses 369,000 of these displaced persons.² Women and children constitute the majority of displaced persons in Somalia, making them increasingly vulnerable to violent attacks and sexual exploitation, often at the hands of security personnel.^2,12

Ensuring access to safe and timely reproductive health care in crisis and conflict/post-conflict settings such as Somalia is critical.¹⁴ The small number of international non-governmental organizations (NGOs) working in Somalia have largely focused on providing services for and responding to the needs of displaced populations. The Federal Government of Somalia, inaugurated in 2012, is the first centralized, internationally recognized government in over 20 years.¹⁵ The country is experiencing a period of relative stability and nation building, and has since re-established the previously defunct Ministry of Health. This provides a unique window of opportunity for interventions addressing women’s reproductive health issues, which have long been coordinated by local and external actors, resulting in considerable fragmentation.^16,17 The safe and timely provision of reproductive health services is not only a human right for all women, but necessary for the development of any country as a whole. Access to family planning, for example, can reduce maternal deaths by up to 40%,¹⁸ and investing in women’s reproductive health interventions leads to reduced levels of poverty and increased productivity and growth.¹⁹ As a consequence, the Federal Government of Somalia has incorporated birth spacing into its national health priorities.¹

Existing security threats in Somalia continue to pose challenges to conducting research on the ground. In recent years, researchers have had to rely primarily on extensive literature reviews and interviews with stakeholders based outside of Somalia to explore health and health-seeking behaviours among Somali populations.^3,6 Other research has focused on reproductive health among Somali populations in Europe and the United States.^20–23 Missing from the literature are the voices of those living within Somalia itself. These perspectives are critical in developing an understanding of women’s reproductive health experiences and health-seeking behaviours, which are integral to the planning and provision of tailored health services in Somalia.³

In the summer of 2014, our research team conducted a multi-methods study dedicated to emergency contraception in Somalia. Our overarching project included key informant interviews, structured in-person interviews with retail pharmacists, and focus group discussions (FGDs) with married and unmarried women in Somalia’s capital of Mogadishu, although none of our FGD participants had ever heard of emergency contraception. We have described the findings from the overall project elsewhere.²⁴ Our FGDs explored a wide range of reproductive health issues. Women discussed their personal experiences, individual and community awareness of reproductive health issues, social norms regarding contraception, abortion, and childbirth, and their opinions about priorities for improving services and programs. In this paper we draw from these FGDs to explore women’s knowledge of, experiences with, and need for reproductive health services.

Methods

Recruitment

We conducted four FGDs with Somali women of reproductive age living in Mogadishu at the time of data collection in 2014. Representatives of our local NGO partner, the Coalition for Grassroots Women Organization (COGWO), recruited participants from several neighbourhoods in the city. We also recruited participants from an internally displaced person (IDP) camp in Mogadishu. Recruitment took place through announcements, word-of-mouth, and personal networks as well as in conjunction with already scheduled home visits by COGWO representatives. Given prevailing social norms and our desire to engage a range of women, we constructed our FGDs around marital status. We conducted three FGDs with married women and one FGD with unmarried women, all of whom came from the IDP camp. Each group was comprised of 5-6 participants.

Data collection

We held our FGDs in a private meeting space provided to our team by COGWO. MY, a Somali national, led the discussions and FG, who is Somali-Canadian, co-facilitated the discussions and took copious notes throughout. We used a discussion guide created specifically for this study and all discussions took place in Somali. The discussions began with women introducing themselves and providing basic demographic information about themselves, their families, and their living arrangements. We then explored women’s knowledge of, attitudes toward, and experiences with emergency contraception. Our next domain of inquiry centered on women’s knowledge of, experiences with, and perceptions of need for other reproductive health services. We focused this section of the discussion on contraception, abortion, pregnancy, and delivery care. We concluded with a discussion of priorities for improving reproductive health services in Mogadishu. We obtained verbal consent from all participants before proceeding with the FGDs and audio-recorded the discussions. Each discussion lasted 60-90 minutes. As a thank you for participating in the study and to cover any costs associated with travel, we gave women USD10 and at the end of the discussion we provided participants with contact information for relevant women’s empowerment organizations and reproductive health agencies in the capital city. Immediately after each discussion, MY and FG debriefed on the facilitation, content, and tenor; FG formally memoed after each FGD in order to reflect further on the overarching dynamics and begin the analytic process.

Analysis

We used ATLAS.ti qualitative data management software to organize our data, which was comprised of translated (into English) transcripts, memos and field notes, and conducted content and thematic analyses. Qualitative data analysis is an iterative, ongoing process that begins with data collection. Based on our study questions and discussion guide we developed a priori (predetermined) codes and as we familiarized ourselves with the data we created additional codes to capture emergent ideas, thus using both deductive and inductive techniques.^25,26 We then engaged in second and third level analyses in order to identify key themes and relationships between ideas and attach meaning and significance to the findings. Because the number of FGDs was small, we used a modified constant comparative analytic strategy in which we conducted pair-wise comparisons of each discussion. This allowed us to further refine our thinking regarding the similarities and differences between the discussion groups, with special attention to marital status and IDP residence. Regular study team meetings guided our overall interpretation.

Participant characteristics

Twenty-one Somali women participated in our FGDs, five unmarried women, aged 18-20, and 16 married Somali women, aged 18-53. All of the married FGD participants and none of the unmarried women had children. Those who were employed at the time of our FGDs owned small businesses (n = 4), worked on farms (n = 1), or worked as nurses and community health programmers (n = 3). Nearly all of the women in our FGDs were ethnic Somalis. None of our married FGD participants relayed information about their levels of education. However, four of our unmarried FGD participants were enrolled in private school in Mogadishu at the time of the discussion.

Our FGD participants came from four different districts in Mogadishu: Abdi-Aziz, Bondhere, Karan, and Shibis. Two of these neighbourhoods, Abdi-Aziz and Shibis, are small in comparison to other Mogadishu districts and characterized by an overall sense of security and stability. In contrast, Karan and Bondhere are two of the largest districts in the city, characterized by higher levels of insecurity and the presence of displaced populations. These districts represent the social, cultural, political, and economic diversity of Mogadishu.

Ethical considerations

The Health Sciences and Sciences Research Ethics Board at the University of Ottawa approved the study protocol (File # H02-14-09). COGWO also reviewed our study and determined that our approach met local research standards. Because of the high rates of illiteracy in Somalia, we asked participants to give verbal consent. In our results section we present the key themes that emerged during our discussions and include illustrative quotes to support our interpretation. We have masked or redacted all personally identifying information and have used pseudonyms for our participants.

Results

Knowledge of contraception is limited and misinformation is widespread

“Only God knows how much I want to space my children apart! I want to space my children apart so bad…right now my child is two months, and my period already came back, what am I supposed to do?” (Noura, 30)

Our participants evinced limited awareness of a full range of contraceptive methods available in Somalia. In our discussions, married women explained that oral contraceptive pills (typically referred to by the brand name “Vente Uno” or “Twenty One”) and condoms could be purchased in pharmacies or procured through some of the NGOs working in the region. However, married women discussed condoms with disdain and repeatedly expressed concern that condoms served as a method to spread HIV. Several women also mentioned Depo Provera but had only vague knowledge of what it is or how it is used. As Amina, a married 26 year old, explained, “There is even an injection for three months. You can get one for one month or three months, and she will take it [i.e. get injected] once”. None of the women in our FGDs referenced long-acting reversible contraceptives and none of our participants had heard of any modality of emergency contraception.

In contrast, almost all of the married women in our FGDs were aware that breastfeeding could be used to promote birth spacing and, unlike other methods of pregnancy prevention, unmarried participants were aware of breastfeeding as well. Indeed, in all four of our discussions women spent more time talking about breastfeeding than all other methods of contraception combined. Most of the married women in our FGDs had used breastfeeding at some point as a means of pregnancy prevention. Two of our participants worked as health programmers in Mogadishu, tasked with visiting women in their homes and encouraging them to space their pregnancies through consistent, effective breastfeeding.

However, married women were generally dissatisfied with breastfeeding for birth spacing and explained that the constancy required for the method to be effective interfered with other responsibilities. As Amina explained,

“When I go to work, the child won’t get proper breast milk, and the period will come…the woman will become pregnant…my child is two years old and now I am already pregnant…and I need to go to work.”

Women engage in a variety of unsafe and/or ineffective abortion practices

“I’m talking about the real honey! If you mix half a litre [of honey with oil] it will release the child.” (Rone, 42)

Our FGD participants described a wide range of practices to terminate a pregnancy. Women repeatedly expressed their belief that abortion is forbidden, both legally and socio-culturally, in Somalia. However, consistent with other settings where abortion is severely legally restricted, women explained that self-induction practices, particularly with Aspirin, anti-malarials and allergy medications, were common. Our participants also detailed a number of traditional practices to terminate a pregnancy. As Amran, age 40, explained, “I have also heard of using the fat from [sheep’s] meat…if you eat that, even if your pregnancy has advanced one month…you will get rid of the baby”.

Women also spoke of xaqitaan, or “sweepers”, who provide abortion services for women in Mogadishu. As Rone explained, “They use the sweeper…they go to a doctor and he sweeps it [the foetus] out”. However, women regarded these providers with scepticism as they were aware of the various risks associated with this unregulated, illicit practice. As 40 year old Farhia shared,

“For example, a woman who after three months realizes that she may be giving birth to a bastard child [will] go to an [abortion provider] and get an injection and then have very bad problems and start bleeding and the child will come out…these women have suffered. This is thievery, someone will say I can bring this child out [just] give me money.”

Mistrust of providers is common and influences care-seeking behaviours

“Somali women first of all do not receive any sort of care in this country. When she becomes pregnant and when she is giving birth are the first times that she sees a doctor…A girl that is the daughter of my aunt went to a doctor when she was four months pregnant and was in pain…[She] got the medications from that gynaecologist [and] the medications that she got killed the child that was inside of her.” (Noura, 30)

“Sweepers” are not the only health care providers that women mistrust. A repeated theme in our discussions was the lack of confidence women had in health service professionals and the care they provide. Married women spoke about the many difficulties they had faced when trying to access health care during their pregnancies. They often described doctors as “fakes”, whose financial motivations superseded the health of the women they were caring for. As Noura explained, “Even when you get pregnant and you say ‘I will visit a doctor’ you have to fear whether or not the doctor might kill your child”. Women explained that doctors practicing in Mogadishu often mistreated women and there were no systems by which to hold them accountable.

Given this overarching context, women in our FGDs rarely interacted with health care providers and generally sought care from women in their communities. Although our participants were generally aware of the risks associated with delivering outside of a health facility, women made decisions to give birth in environments that they were comfortable with and with attendants that they trusted. Women in our FGDs relayed harrowing ordeals of births in hospital settings. However, it was clear that measures were being taken in Mogadishu to encourage women to give birth with the assistance of a trained health professional, and to ensure women were not seeking medical care from unqualified or untrained medical providers. As Farhia, a community health worker explained, “If she is pregnant…[we] make sure that she doesn’t take medications from unreliable doctors”.

In addition, our FGD participants highlighted costs as a major barrier to accessing existing reproductive health services in Mogadishu. The overwhelming majority of hospitals and clinics in the capital are privately owned and operated and women explained that the fees attached to services were prohibitive and further undermined their trust in health service professionals. As Rone explained, “The doctors here are not like the doctors from before [the war] - they just open up clinics…to feed their children like everybody else”. Although women were aware of free care offered by both local and international NGOs, they explained that these services were typically reserved for women in IDP settlements or survivors of rape. Our participants identified safe and affordable pregnancy-related care as a top priority.

The need for culturally resonant reproductive health information and services is acute

“There needs to be a lot of awareness-raising and education done, people need to be reached and to be talked to. It will not take one day or two days or a year for these people to understand, it will take a lot of time.” (Ayne, 40)

Women in all of our FGDs spoke often about the need for awareness-raising campaigns and information. There was general agreement that information about a wide range of reproductive health issues is necessary and that targeting both married women and men should be prioritized. Somali society is organized along patriarchal lines, and married women explained that many women require their husband’s approval to adopt a method of contraception. Excluding men from awareness-raising campaigns regarding matters such as contraception can have dire consequences, as explained by Naima:

“I have seen my neighbour who had 12 children, her youngest child was a bit paralyzed, and she said she couldn't afford to have any other children… so she said she was going to get injected [with Depo Provera]. He [her husband] refused. She was stubborn so she went and got injected anyway, so he [divorced her]; she supports her children on her own now.”

Women were enthusiastic about promoting information and services to support birth spacing. They explained that rapid repeat pregnancies took a toll on women, their children, and their families. Farhia drew from her personal experience of having six children and explained,

“[My children] will not find breakfast, lunch, and dinner… just pancakes, and all of the children develop anaemia. No meat, no eggs, no liver - they are just eating pancakes that are watery [with] some coloured tea, that’s it. So why wouldn't the children develop anaemia?”

Women who were the sole breadwinner in the family described the financial hardship that came with having another pregnancy while caring for an infant. For these women, the ability to space births and therefore have the option of continuing to work was of paramount importance.

The concept of birth spacing was also perceived by both our married and unmarried participants as both culturally and religiously resonant. As unmarried Anab summarized, “It is good to space the children. It is good for their health, and even for the religion because the children will grow up properly”. Married women differentiated “birth spacing” from “family planning”, perceiving the latter as limiting or restricting the number of children. Within this frame, women explained that oral contraceptive pills, Depo Provera, and emergency contraception (once we had provided extensive information about the dedicated progestin-only pill), were culturally and religiously acceptable when used by a married woman. Our participants explained that awareness-raising campaigns and services would be most successful and acceptable if they were positioned around birth spacing. There is a long pro-natalist tradition in Somalia and large numbers of children are generally viewed as a source of socio-cultural prosperity, particularly for men. As Noura stated, “They [men] want that you give birth every year… [and the] women do not want their husband to marry another women, so that is why [they comply].”

Discussion

The findings from our FGDs reveal that there is much to be done to improve reproductive health information, services, and outcomes in Somalia. Our results echo those of previous studies exploring Somali women’s reproductive health needs, albeit in contexts outside of Somalia’s borders^3,6,27 and suggest that the need for action is pressing. Allowing for opportunities for stakeholders in Somalia to engage with stakeholders in neighbouring countries may prove beneficial for future reproductive health interventions in the country. Countries like Ethiopia and Kenya, which have strong national family planning strategies in place^28,29 and large native and refugee Somali populations, may provide a wealth of expertise and knowledge for stakeholders within Somalia. Thus identifying and facilitating intra-regional exchange appears warranted.

Our participants’ reproductive health experiences and those of their family members, friends, and other community members were shaped by a variety of factors, including misinformation, restrictive laws and policies, mistrust of clinicians, and prohibitively expensive services. Our findings reveal that the low contraceptive prevalence rate in Somalia is likely not completely attributable to lack of awareness, misinformation, or cultural-religious opposition. Indeed, consistent with theological and juridical positioning within Sunni Islam^30–32 our participants were clear that non-permanent methods of contraception were permissible when used by a married couple. And women in our FGDs evinced some awareness of available contraceptive methods, even if they had never used these methods themselves. Although there is certainly a need for evidence-based awareness-raising campaigns, these efforts are unlikely to sufficiently address the other dynamics that have resulted in a contraceptive prevalence rate of 1%.⁵

When developing reproductive health programs, women’s concerns about safety, cost, and service quality, which were highlighted throughout our FGDs, must also be considered. Studies in other parts of sub-Saharan Africa have demonstrated that cost often serves as a major barrier to accessing reproductive health services.¹⁰ High rates of extreme poverty in Somalia mean that many women are unable to provide basic necessities for themselves and their children.³³ Approximately three-quarters of the population currently live in poverty, and the annual per capita income is US$284.³⁴ As a consequence, women reported that they prioritized other issues – including food, water, and safe shelter – over health care. Recognition of this dynamic and the competing priorities women face will need to be built into awareness-raising campaigns in order for them to be successful. Further, the emergence of unregulated private sector facilities and providers in the post-conflict period has contributed to mistrust of both clinicians and the quality of the services provided. The establishment of a national health system may provide a window of opportunity to support the training and accreditation of health service professionals, create national standards and guidelines, and establish systems of accountability.

In recent years, primacy has been placed on increasing awareness about breastfeeding as a method of pregnancy prevention and the small number of programs that have been launched in Mogadishu reflect this orientation. [Representative from the Ministry of Health, Personal Communication, May 2014] However, the experience of our participants with lactational amenorrhea was decidedly mixed and women identified a number of familial and economic challenges to being able to effectively employ this method of pregnancy prevention. Future efforts to engage with women in their homes and through community organizations may benefit from integrating breastfeeding information with a broader array of birth spacing strategies and contraceptive methods. Our participants were very clear about the cultural resonance of messages related to birth spacing and thus efforts to expand reproductive health services would likely benefit from incorporation of this frame.

In 2010, the fledgling federal government developed the Somali Reproductive Health National Strategy and Action Plan. This five-year action plan included birth spacing, safe delivery, and the prevention of harmful practices among its top priorities.¹ Although few efforts have been undertaken in Somalia to advance these aims, the existence of the Somali Reproductive Health National Strategy and Action Plan signals that reproductive health is on the national agenda for the first time in decades. Further, decision-makers in a variety of sectors have recognized the importance of health, in general, and reproductive health, in particular, to the advancement of broader development goals. As the Federal Government of Somalia moves forward with the creation of a new five-year plan, there is a window of opportunity to engage a variety of stakeholders in discussion about reproductive health priorities. As Somalia begins rebuilding state institutions, such as the Ministry of Health, after over 20 years of conflict and mass displacement, it is important that women’s concerns are included in decision-making efforts shaping the development of future women’s health and reproductive health policies.

Limitations

Our study has a number of limitations. Although we are confident that the themes that we identified have import beyond the immediate study population and reflect broader social norms, the qualitative nature of this study by definition means that our findings are not generalizable. Furthermore, for a number of pragmatic reasons, including the security situation at the time of data collection, we limited our project to Mogadishu. Services are more widely available in the capital and many of the service delivery NGOs concentrate their work in this region. Future projects would benefit from including women in other regions of the country.

Two investigators co-led our FGDs; one a women’s rights activist and community organizer from Mogadishu, the other a Canadian master’s student of Somali descent. As is true of qualitative research in general, the positionality of our study team members, including national origin, gender, educational level, marital status, and dialect employed, undoubtedly influenced the FGD interactions. Through debriefings, regular team meetings, and formal memos, we believe that we were able to understand these influences and enhance the credibility and trustworthiness³⁵ of the study.

Conclusion

Women living in Somalia have some of the highest levels of unmet reproductive health needs in the world. Restrictive laws and policies, misinformation, mistrust of clinicians, and high cost of services shape women’s experiences. Avenues to expand culturally resonant information and services need to be identified and incorporating women’s perspectives into these efforts is critical. As Somalia embarks on a period of rebuilding and recovery, reproductive health merits a place on the national agenda.

Acknowledgements

This project was supported by a trainee grant from the Society of Family Planning (FG). AF’s Endowed Chair is funded by the Ministry of Health and Long-Term Care in Ontario and we appreciate the general support for her time that made this project possible. The study team thanks COGWO for their support of the project and assistance with recruitment, facilitation, and translation. Finally, we thank Drs. Shoshana Magnet and Sanni Yaya for their feedback on earlier phases of this project. The conclusions and opinions expressed in this article are those of the authors and do not necessarily represent the views of the organizations with which the authors are affiliated or the funders.