Background: COVID-19's emergence questions the agility of health policy deployment in a context of urgency. This exceptional pandemic offers a unique Implementation Science study opportunity. It reveals how actors adapt, coordinate, and mitigate an unknown global threat to safeguard populations from an initially mysterious virus. Limited research has explored how involved players act and adapt their practices to fulfil health protection missions during a global health crisis. Bridging the gap between public policy expectations and achievements requires a methodology for stakeholder identification and implementation practice description. Objective: Focusing on COVID-19 management in France's second-largest region, we investigate ministerial recommendation implementation and the emergence of new links, coordination modes, and practices. Methods: Due to the novel subject, we adopted grounded theory. Initial documentary data collection identifies stakeholders for subsequent interviews. Open-ended coding of collected discourse enables content analysis. Results: Findings reveal a crisis-driven re-evaluation of stakeholder relationships. This research identifies three levels of implementation of health policies at the local level (administrative, organizational and operational) and reveals different types of coordination specific to each of these levels. Our results provide insights on how to better coordinate and implement healthcare policies in a period of crisis. Recommendations include real-life simulations of large-scale crises. Conclusion: Our work establishes a methodological foundation for analyzing coordination dynamics. Future research could compare these findings with other unpredictable health emergencies, such as episiotomic veterinary health crises.  The first step of the method is to analyse the guidelines of health policy implementation during the Covid-19 crisis and to identify the main stakeholders in charge of the local health policy implementation.  The second step consists of interviewing these stakeholders using a co-constructed sample and structural coding of their speech to reveal the forms of coordination between stakeholders.