Using the implementation research logic model (IRLM) to track implementation strategy changes and adaptations: a case example from a study on APOL1 genetic testing and counseling by nephrologists

The Implementation Research Logic Model (IRLM) was developed to help implementation researchers specify relationships among determinants, implementation strategies, mechanisms, and outcomes. Although the IRLM is intended to support planning, execution, reporting, and synthesis, it is often treated as a static product used for one purpose—most commonly for planning. Even when used across these purposes, a reporting gap may arise when strategies are adapted, bundled, discontinued, or implemented differently across sites. However, the IRLM can be used to execute and track the study or implementation effort. We describe a structured method for using the IRLM across three versions and outputs: (1) a protocol IRLM that represents the study’s theory-of-intent; (2) a tracked revision of the IRLM that documents changes, rationales, and evidence sources for adaptations and modifications; and (3) a final reconciled IRLM that represents the enacted theory-in-use. We illustrate the use of this methodology with the Apolipoprotein L1 ( APOL1 ) Counseling and Testing Program, a hybrid type 2 effectiveness-implementation study using a two-site non-randomized trial of culturally competent APOL1 genetic testing, a conversational agent (chatbot), and nephrologist-delivered counseling for living kidney donor evaluation. Applying the versioned IRLM method to the APOL1 trial generated a strategy execution matrix and a revised set of context-strategy-mechanism-outcome linkages. The process made visible several forms of protocol-to-practice change including: discontinued electronic health record (EHR) integration strategies, adapted conversational agent and educational brochure delivery, site-specific task shifting, bundled stakeholder engagement strategies, and modifications to nephrologist training supports. Mechanism reassessment led to adding donor empowerment, separating communication from coordination, and removing an EHR-related complex mechanism that was no longer actionable. Treating the IRLM as a dynamic implementation product can improve transparency about what happened during implementation trials, why actions taken differed from the protocol, and how those differences alter hypothesized causal pathways and outcomes. This method complements longitudinal strategy tracking approaches and can support reporting, interpretability, replication, scale-up, and mechanism-focused implementation science.