Mental health information systems in low-resource settings are scarce worldwide. Data collection was accurate, yet sustainable staffing was a challenge when using task-shared clinical providers for data collection in health centers in rural Haiti. Integrating mental health data collection within existing data collection systems would help close this key gap., Key Findings Balancing time for clinical duties with time for data collection was a key challenge when using task-shared health care workers to collect data.Feedback of data to users, ongoing supervision, and opportunities for rewards for high-performers may help increase provider buy-in for new data collection systems.Retaining paper forms alongside digital data collection systems or the inclusion of decision support tools directly within digital data collection systems may be needed to fully support health care worker learning in a task-shared mental health system. Key Implications Program managers should carefully consider sustainable staffing from the beginning when designing digital data collection projects.Program managers should consider how digital data collection systems can incorporate decision support tools to sustainably support ongoing learning of task-shared providers.Policy makers should integrate mental health data collection within other data systems to leverage infrastructure and resources for mental health., Introduction: Effective digital health management information systems (HMIS) support health data validity, which enables health care teams to make programmatic decisions and country-level decision making in support of international development targets. In 2015, mental health was included within the Sustainable Development Goals, yet there are few applications of HMIS of any type in the practice of mental health care in resource-limited settings. Zanmi Lasante (ZL), one of the largest providers of mental health care in Haiti, developed a digital data collection system for mental health across 11 public rural health facilities. Program Intervention: We describe the development, implementation, and evaluation of the digital system for mental health data collection at ZL. To evaluate system reliability, we assessed the number of missing monthly reports. To evaluate data validity, we calculated concordance between the digital system and paper charts at 2 facilities. To evaluate the system's ability to inform decision making, we specified and then calculated 4 priority indicators. Results: The digital system was missing 5 of 143 monthly reports across all facilities and had 74.3% (55/74) and 98% (49/50) concordance with paper charts. It was possible to calculate all 4 indicators, which led to programmatic changes in 2 cases. In response to implementation challenges, it was necessary to use strategies to increase provider buy-in and ultimately to introduce dedicated data clerks to keep pace with data collection and protect time for clinical work. Lessons Learned: While demonstrating the potential of collecting mental health data digitally in a low-resource rural setting, we found that it was necessary to consider the ongoing roles of paper records alongside digital data collection. We also identified the challenge of balancing clinical and data collection responsibilities among a limited staff. Ongoing work is needed to develop truly sustainable and scalable models for mental health data collection in resource-limited settings.