"Koyi Sawaal Nahi Hai": Reimagining Maternal Health Chatbots for Collective, Culturally Grounded Care

TL;DR

The paper analyzes how maternal-health chatbots operate in fragile, collectivist settings by deploying a WhatsApp-based system with 48 pregnant women in Lahore, revealing that adoption hinges on mediated decision-making, shared phone use, and infrastructural fragility rather than interface usability. It introduces the Relational Chatbot Design Grammar (RCDG) comprising four commitments—Mediated Decision-Making, Silence and Endurance, Episodic Use, and Fragile Contexts—to guide culturally grounded, collective care. Through phase-wise qualitative methods and triangulation with clinician insights, the study shows that trust, legitimacy, and engagement are distributed across family networks and health-system actors, framing non-use as a relational pattern rather than individual failure. The findings urge designers to embed resilience, relational authority, and context-sensitive interactions into chatbots to sustain care in LMIC health systems where decisions are negotiated and access is episodic.

Abstract

In recent years, LLM-based maternal health chatbots have been widely deployed in low-resource settings, but they often ignore real-world contexts where women may not own phones, have limited literacy, and share decision-making within families. Through the deployment of a WhatsApp-based maternal health chatbot with 48 pregnant women in Lahore, Pakistan, we examine barriers to use in populations where phones are shared, decision-making is collective, and literacy varies. We complement this with focus group discussions with obstetric clinicians. Our findings reveal how adoption is shaped by proxy consent and family mediation, intermittent phone access, silence around asking questions, infrastructural breakdowns, and contested authority. We frame barriers to non-use as culturally conditioned rather than individual choices, and introduce the Relational Chatbot Design Grammar (RCDG): four commitments that enable mediated decision-making, recognize silence as engagement, support episodic use, and treat fragility as baseline to reorient maternal health chatbots toward culturally grounded, collective care.

Figures (4)

• Figure 1: Study phases. The study unfolded across four sequential phases. Phase 0 established context and recruitment, Phase 1 piloted and then pivoted the deployment strategy, Phase 2 gathered provider perspectives, and Phase 3 synthesized data through analysis and triangulation. Solid arrows show the planned sequence of phases, while the dotted arrow marks the pivot from the initial deployment attempt.
• Figure 2: Pathways of pregnancy care negotiations. Flow of symptoms, decision-making, and advice across individual, household, and clinical levels, mediated by family, social norms, and parallel health authorities. The chat bot enters this circulation; its guidance has to be legitimated or contested along with these existing negotiations.
• Figure 3: Pyramid of relational authority in maternal health chatbot use. Doctors anchored authority, kin mediated consent, and chatbots were situational and relationally framed, showing adoption as collective rather than individual.
• Figure 4: Nested dynamics of maternal health chatbot use. The figure illustrates micro-level embodied experiences, meso-level household dynamics, and macro-level structural conditions intersect to shape how women engage with chatbots. This ecological view grounds our Relational Chatbot Design Grammar, framing silence, family mediation, episodic use, and fragility as key design commitments.