4HerPower Challenge: Innovating for Sexual and Reproductive Health and Rights
Línea Zona Joven: exploring without reserve
Provide a one-line summary of your solution.
Development of an AI model integrated to a virtual youth care channel focused on resolving doubts about sexuality in a friendly, straighforward and assertive way for the promotion of contraceptive measures and prevention of early parenthood
Film your elevator pitch.
What is your solution?
Línea Zona Joven is a specialized service hotline, operating through a WhatsApp chat designed exclusively for adolescents and young people. Its primary objective is to provide immediate solutions to the needs for information on sexuality, based on the principles of confidentiality, reliability and excellence in the answers provided to each query. Currently, more than 26,000 adolescents and young people between the ages of 10 and 28 have had their doubts addressed, from June 2022 to September 2023.
The solution is implemented in an open and cost-free channel provided by Profamilia, operating as a prototype of the "Wizard of Oz", in which two people attend to the queries of adolescents and young people. From the technological angle, we have a multi-agent platform that serves as the basis for managing all the chats. On this platform, the flow of conversations is managed and information is collected for further analysis and decision making related to the next steps of the hotline.
For users, the experience is similar to having a WhatsApp conversation with a friend. Operationally, support is provided from the youth-friendly health services area, where technical guidance is provided to those answering questions. Once the doubt is resolved, users can be referred to schedule appointments with Profamilia. This ensures that they not only have access to information, but also receive comprehensive care that includes sexual and reproductive health services.
What we are looking for with this project is to strengthen our solution by integrating a natural language artificial intelligence (AI) model into the care and assistance process. This next stage of the prototype has the following objectives: i) Develop a robust natural language model capable of providing adequate answers to frequently asked questions posed by young people through the youth hotline, thus ensuring the availability and access to information on sexuality. Ii) Implement the solution in our network of RedJoven volunteers, once the hypotheses of fluency, quality and evaluation of the model have been validated. III) Significantly increase the quality and decrease the response time to improve access to information for those who consult.
Finally, with the implementation of this AI, our goal is to provide clear and simple explanations about contraception, which constitutes 30% of the consultations made by adolescents and young people. Questions such as "What are contraceptive methods?", "How are they used?", "How can I access them?", "What is the best method?" and "How can I take care of myself?" will receive timely answers, which will allow us to provide information and strengthen the well-informed decision-making process about their sexual and reproductive health and contribute to: I) Decrease the number of unwanted pregnancies, which in turn can have a positive impact on the social and economic development of the country. II) Reduce the number of unsafe abortions, which can have serious consequences for women's health. III) Prevent the transmission of STIs, which can cause serious health problems such as infertility, HIV/AIDS and/or cancer.
What specific problem are you solving?
During adolescence and youth stages people face multiple vulnerabilities and risks regarding their sexual and reproductive lives that expose them to problematic events such as adolescent pregnancy, different forms of gender-based and sexual violence, or STIs (including HIV). Besides, it has become evident that young people face significant challenges when trying to obtain clear and precise answers to their concerns about SRH, which partially explains their lack of information about aspects such as family planning or early parenthood.
In Colombia, according to DANE figures for 2022, 18% of reported pregnancies were considered teenage pregnancies. Out of this 18%, 0.8% were pregnancies of girls between 10 and 14 years old, while the rest are attributed to girls between 15-19. Therefore, despite a reduction of 7.1% compared to the previous year, this figure is consistent with the regional average, which itself holds the second-highest rate of teenage pregnancies in the world (UNFPA, 2023).
On the other hand, a research on how the different audiences in the country behave regarding their SRHR was presented by Profamilia in 2023. The majority of the people involved (51%) were young people between 18 and 23 years old, and 25.6% mentioned that no one advises them when they have questions related to their SRH. On the other hand, 17.8% said that they prefer to go to health professionals (clinics, hospitals, or health centers) for counseling, while 14% said that they seek counseling through a group or collective of young people, and in a smaller proportion (8%) seek counseling from private health professionals.
In other findings, it was identified that the consultation sources are mostly Internet pages such as Google, Yahoo, Bring (74%). Another 39% of people surveyed get advice through social networks such as WhatsApp, Instagram, Facebook or Tik Tok. Finally, the study shows that the most requested topics include contraceptive methods (46.7%), care practices (23%), VTP (15%), and STI prevention or testing in 8%. There is a low participation in topics such as diverse sexuality, abuse, and sexual violence.
In this sense, after revealing some findings on SRH consultation and exploring teenage pregnancy rates in the country, it is evident that the lack of reliable and accessible sources of information, as well as embarrassment or inhibitions to raise these concerns to their partners, family members or health professionals, generate a gap in young people's access to services and knowledge essential for their physical and emotional wellbeing. Also, a lack of care and guidance services targeted specifically at young people has been identified, focusing on aspects such as contraception, prevention of STIs, sexual and gender diversity, and mental health.
This problem is exacerbated by the growing influence of technology and social networks, where young people seek information but find themselves exposed to unverified content that is potentially harmful to their health and well-being. In addition, health institutions and entities have not been able to effectively meet this need, either due to lack of resources, outdated approaches, or lack of effective and close communication with young people.
Who does your solution serve, and in what ways will the solution impact their lives?
In order to have a better understanding of the needs, concerns and dissatisfactions of adolescents and young people, 38 semi-structured interviews were conducted at the beginning of the project. From these, the following insights were identified: I) Young people consult about SRH mainly on the Internet, but the information they find may not be reliable. II) Young people are embarrassed to consult their partners, family members or professionals about their doubts. III) There is a lack of information and guidance on sexual and reproductive health, immediately and by experts. IV) Health Promoting Entities do not provide information on sexual and reproductive health, or do so inadequately.
Based on these insights and a study of trends, we created the Línea Zona Joven. With it, we validated under the Lean Startup methodology that, in the first iterations of the prototype that a WhatsApp chat allowed us to be close and give truthful answers solving the embarrassment and lack of confidence in the sources used on the internet.
Although the Línea Zona Joven is a solution aimed at adolescents and young people at the national level, for this stage of the project, we worked with and for the RedJoven de Profamilia, a volunteer network that integrates more than 1000 youth and adolescent leaders from various regions of the country; which works to strengthen skills, knowledge and attitudes, for the development of autonomy and empowerment of Sexual and Reproductive Rights in Colombia. RedJoven has positioned itself as one of the largest youth platforms in Latin America that has an impact through its actions towards peers, encouraging informed decision making in SRH.
For this solution, it is intended to articulate, initially, with the Health with Youth Glasses Node, as a subcommittee of RedJoven made up of 59 young people who focus their work on friendly health services for adolescents and youth, as well as with the Training Node, made up of 47 young people interested in positioning contents and messages on Comprehensive Sexuality Education for the young audience; these teams will be able to accompany the development of the initiative, as well as the validation of the results from their context and youth experience. Additionally, young people will be able to count on a line of support as leaders, obtaining clear and accessible information, even in the midst of replication actions in diverse communities and territories.
During the first year of implementation of the hotline, it’s been observed that 80% of the users accessing the service are women, while 19% are men, and 1% are people with trans and/or non-binary life experiences. Regarding the development of AI, we will work with the intersectional approach to limit biases and stereotypes that may reinforce discrimination to the person based on gender, race, ethnicity, sexual orientation or disability. To this end, we intend to recognize inclusive language and promote gender equity, meaning that we will not rely on responses that may increase the gaps and inequalities in SRH of men, women and people with diverse life experiences.
How are you and your team well-positioned to deliver this solution?
The team leading this solution is made up of 4 people, two young collaborators of Profamilia and two young representatives of the RedJoven nodes.
First of all, the leader is Gabrielle Devia Vasquez, a young transman of 26 years old, passionate about social innovation. Currently Innovation Coordinator of Profamilia, who, based on him experience and knowledge in the development of products and services with a social focus, will be in charge of taking the solution to its effective implementation. In addition, he will bring an intersectional perspective due to the different community processes in which he has participated specifically in the trans community.
Secondly, the team is supported by the participation of Profamilia's National Youth Coordination Alexandra Sanchez, who lead actions related to Adolescent and Youth Friendly Health Services, and accompany the development of youth volunteering within the Organization, in addition to previously integrating the RedJoven as young volunteers.
Thirdly, the team is composed of two leaders who are part of Profamilia's YouthNetwork, who have been democratically delegated to coordinate the National Nodes of Training (Tatiana Ocampo-psychology student) and Health with Youth Glasses (Carlos Castrillón-nursing student ). Their leadership is fundamental for youth representation and the positioning of their positions in decision-making spaces, as well as their effective integration in differential strategies for young people.
The participation of the team mentioned above guarantees joint and articulated work with the young beneficiary population, always taking into account their interests, needs and initiatives, and integrating them into the strategy.
It is worth noting that the young people from the nodes that integrate the team are representatives of their communities who have consolidated themselves as youth leaders due to the empowerment and advocacy they have developed in the field of Sexual and Reproductive Health by being part of Profamilia's RedJoven (Youth Network). These young people know firsthand the needs of young people in their territories, they know what their requirements, demands, requests, demands, desires and wishes are; and to that extent they allow us to integrate these visions of the communities into our solutions and thus create contextualized actions that respond to the particularities of the territories and the intersections that affect young people.
Finally, we consider it important to emphasize that from the process of designing our solution we have sought to ensure that it responds to the needs of young people and that their perspective is transversally represented in each step of its development and growth. For this reason, since the beginning we carried out the interview process where we gathered the needs of young people within the RedJoven to consolidate the solution, and now, we contemplate the participation of some of their representatives for its growth stage.
Which dimension of the Challenge does your solution most closely address?
Strengthen the capacity and engagement of young innovators in the development, implementation and growth of solutions addressing their SRHR needs.In what city, town, or region is your solution team headquartered?
Bogotá, ColombiaIn what country is your solution team headquartered?
What is your solution’s stage of development?
Growth: An organization with an established product, service, or business model that is rolled out in one or more communities
How many people does your solution currently serve?
To date, 38,000 people have used the line; however, it should be noted that 79% are adolescents and young people, so that an average of about 26,000 young people have used the line.
Who is the Team Lead for your solution?
Gabrielle Devia Vázquez
What makes your solution innovative?
This initiative is innovative since it allows us to design a bot whose main objective is the development of an intelligent and accurate chatbot to answer users' questions about sexuality and reproduction. This bot will have an essential component: 1) Answers to frequently asked questions. Our approach will be to use open source Generative Language Models (LLMs) for the implementation of this AI. The specific model to be used has not been defined at this time, as this decision will depend on its availability and suitability for our non-commercial purposes.
The process of building the AI will involve certain steps. After selecting the appropriate LLM model, we will focus on customizing it to suit our specific needs. To do this, we will carry out a fine-tuning process of the model, using Profamilia's existing documents as a basis. This phase will allow the model to understand and adapt to the specific linguistic and thematic patterns present in our data.
The data base is a fundamental aspect for learning the model. The idea is that the modeled chatbot ingests the information available in the provided documents and uses it knowledge for its conversational and learning capabilities. This data is of vital importance as it will allow our chatbot to understand and reproduce the appropriate terminology, style and tone for the conversations it will have with users.
It is important to understand that it is not necessary to pre-train the algorithm from scratch, as we will be using a pre-trained model. The idea is to apply a re-training or 'fine-tuning' to that model, using our specific data. This will allow considerable savings in time and resources, while ensuring a high level of accuracy and adaptability.
Regarding chatbot management, it is mentioned that we are not proposing an administration module to incorporate new user questions. At this time, our goal is for the chatbot to autonomously manage conversations with users. If more data needs to be added to the model, this would involve a retraining process. However, we are willing to consider including an administration module in the proposal if it is decided that it is needed in the future.
In addition, it is emphasized that the option to transfer the conversation to a human agent will always be enabled in case the user requires more personalized attention. In this regard, filters and verification mechanisms will be implemented to ensure that the answers provided are accurate, reliable and supported by verified sources. The bot's knowledge base will be constantly updated and reviewed to ensure the quality of the information.
Therefore, the proposed solutions focus on: Natural Language processing and generation and conversational systems for patients, audio and signal processing, recommendation systematization and grouping of people and causal inference. On the other hand, the methods to be implemented focus on: supervised and unsupervised algorithms, reinforcement learning algorithm-Matching Learnig and generative algorithms.
What are your impact goals for the next year and the next five years, and how will you achieve them?
The project specifically contributes to the Sustainable Development Goals:
SDG 4. Quality Education: The Youth Zone Line accurate and reliable information on contraceptive methods, empowering users to make informed decisions about their sexual and reproductive health. This promotes quality education by providing essential knowledge for responsible decision-making and the prevention of unwanted pregnancies.
SDG 5. Gender equality: By providing women and girls with access to information on contraceptive methods, they are empowered to exercise their right to decide and exercise control over their reproductive health. This is fundamental to promoting gender equality by enabling women to make decisions about if and when they wish to have children.
SDG 10. Reducing inequalities. We seek to contribute to reducing inequalities by providing information and support to people who live in rural, marginalized communities or who have difficulty accessing sexual and reproductive health services. Answering questions and providing information about contraceptive methods helps to reduce disparities in the incidence of unintended pregnancies. This is especially important in populations that are vulnerable or have less access to health services.
To measure this impact that we seek to generate with the Youth Zone Line in terms of access to sexual and reproductive health information, specifically in anti-conception, is to achieve:
- By 2028 more than 200,000 adolescents and young people nationwide who have resolved their doubts.
- At least 10% of these adolescents and young people are able to access a sexual and reproductive health service, thanks to the information provided.
- 50% of the adolescents and young people served must be in intermediate cities and rural areas.
- 50% of the adolescents and young people served should be women and people with diverse life experiences.
This projected impact is based on the data that we currently measure on the line, we seek to strengthen access to health services once the information is acquired, maintain the intersectional approach by resolving doubts of women, diverse people and people in rural areas who have more need for information.
AI will strengthen the information provided by the human agents, that is, we seek to improve the quality in three indicators:
- Response time
- Clarity of the response provided
- Number of responses given in a period of time.
Describe in simple terms how and why you expect your solution to have an impact on the problem.
esults:
From the realization of this project and the development of the model in question is expected to have a model of a Chatbot designed from AI that allows to answer the most frequently asked questions posed by users of the Lina Zona Joven as a channel of attention of Profamilia that allows to have a direct impact on the lives of young people in terms of contraception and the prevention of early parenthood in the adolescent period.
Impacts:
Quality of service. With a greater capacity for real-time response, we will be able to serve a greater number of users throughout the country. But not only that, we will also be able to provide personalized responses tailored to the specific needs of each individual. This will enable us to offer a quality service that is tailored to the particular needs of each user, thus improving their experience and satisfaction.
Expanding service to more users: By incorporating this technology and increasing our response capacity, we aim to reach an even larger number of users. Our focus will be particularly on the cities in between, where, despite existing levels of connectivity, access to sexuality information may be limited. We want to ensure that these users also have access to the information and support they need, providing them with an inclusive and equitable service.
Expansion into new segments: Once we have consolidated and refined our dedicated youth pipeline, we have ambitions to expand our services to encompass other key stakeholders in youth sexual and reproductive health education and information. This includes parents, caregivers, teachers and health personnel. By extending our attention to these groups, we will be able to provide them with resources and appropriate guidance, thus fostering a comprehensive supportive environment for young people and strengthening the impact of our work in society.
If your solution has a website or an app, provide the links here:
https://wa.me/+573185310121 (Opc 2) or https://www.instagram.com/reel/CydesZYs2Fo/?utm_source=ig_web_copy_link&igshid=MzRlODBiNWFlZA==
In which countries do you currently operate?
In which countries will you be operating within the next year?
What type of organization is your solution team?
Nonprofit
How many people work on your solution team?
4 people are presenting the solution:
-Gabrielle Devia (Innovation coordinator)
-Alexandra Sanchez (Young Analyst)
-Tatiana Ocampo (Psychology Student- Training Node representative
-Carlos Castrillón (Nursing Student-Health with Youth Glasses representative)
How long have you been working on your solution?
This solution started in June 2022, which means that to date, our solution has been working for 1 year and four months.
What is your approach to incorporating diversity, equity, and inclusivity into your work?
ender, equity and inclusion considerations:
During the first year of implementation of the Hotline, it has been observed that 80% of users accessing the service are women, while 19% correspond to men, and 1% are made up of people with trans and/or non-binary life experiences. Therefore, it is essential to have approaches that effectively address the needs of each of these populations. For this project, we will work with the rights-based approach, the gender approach and the intersectional approach. Therefore, the model must be developed without discriminating against any person on the basis of gender, race, ethnicity, sexual orientation or disability. Biases and stereotypes that may reinforce discrimination shall be avoided.
These approaches are key elements of the integrality of comprehensive sexuality education that puts into practice the exercise of human rights. Therefore, from the integration of these approaches, we intend to eliminate gender biases and stereotypes when answering young people's questions, recognize inclusive language and promote gender equity, that is, not to fall into answers that may increase the gaps and inequalities in sexual and reproductive health of men, women and people with diverse life experiences.
To promote such approaches, the training dataset will contain information with gender perspectives. The possibility of constructing a set of rules for the model that regulate the way it generates text will be explored. These rules could include avoiding the use of sexist or stereotypical language. Finally, the performance of the model will be monitored and changes made as needed. This will help ensure that the model continues to generate text that is inclusive and representative of diverse human experiences.
Solution Team
-
GD
Gabrielle Devia Vázquez Asociación Profamilia
to Top
What is the name of your solution?
Línea Zona Joven: exploring without reserve