Co-designing and optimising a primary care implementation model for childhood pneumonia management in a low-resource setting using principles of implementation research

Background: Pneumonia remains the leading infectious cause of under-five mortality, particularly in low- and middle-income countries (LMICs) where translation of evidence-based guidelines into practice faces systemic barriers. Although India launched the Childhood Pneumonia Management Guidelines (CPMG), effective implementation remains suboptimal. This study demonstrates how implementation research can support the systematic development and optimisation of a primary care--focused implementation model for childhood pneumonia management in a low-resource setting. Methods: This manuscript reports findings from Phases I and II of a 26-month, pre-post quasi-experimental implementation research project conducted in Palwal district, Haryana, India. The Phase I is formative research including a baseline survey (June-August 2023); Phase II: development and iterative optimisation of the implementation model through co-design within a learning cluster (September 2023-May 2024); and Phase III: implementation and evaluation of the optimised model followed by district-wide scale-up (May 2024-March 2025). Phase I (June-August 2023) employed the Consolidated Framework for Implementation Research (CFIR) to systematically identify multilevel determinants affecting CPMG implementation across a study area of approximately 108,000 inhabitants. Phase II (September 2023-May 2024) focused on co-designing context-specific implementation strategies using the CFIR-ERIC (Consolidated Framework for Implementation Research-Expert Recommendations for Implementing Change) Barrier Busting Tool and iteratively refining an implementation model through participatory co-design workshops with government and community stakeholders within a learning block of approximately 50,000 inhabitants. An optimised Implementation Research Logic Model (IRLM) was developed to align determinants, strategies, mechanisms of change, and implementation outcomes. Results: Thirty-three tailored implementation strategies were identified, addressing barriers across inner (e.g. implementation climate), outer (e.g., socio-cultural norms), individual (e.g. skills), process (e.g., planning), and innovation (e.g., complexity) CFIR domains. Iterative refinement through three successive models (Model 0+, Model 1, and Model 2) within the learning block was associated with improvements in key outcomes: care-seeking from appropriate primary care facilities increased from 0.8% at baseline to over 76% [193/254 cases], appropriate diagnosis improved from 0% to 92.7% [179/193 cases], and fidelity to guideline-based management reached 86% [154/179 cases] by the end of Phase II. These improvements informed the finalised IRLM which was subsequently implemented and evaluated in Phase-III. Conclusions: This study demonstrates a systematic, transparent, and participatory approach to developing and optimising an implementation model for primary care-based childhood pneumonia management in a resource-constrained setting. The co-designed IRLM, grounded in established implementation research frameworks, offers a replicable process for translating evidence-based guidelines into primary care practice in similar LMIC settings. Phase III along with the district-wide implementation and evaluation findings will be reported in a forthcoming paper. Trial Registration: Clinical trial registry CTRI/2021/03/031622 [date: 01/03/2021].