Background: The Maternal, Infant, and Early Childhood Home Visiting (MIECHV) program is a federal public health initiative which supports at-risk families through evidence-based programs and promising approaches for pregnant women, and childhood development for children aged 0 to 5. These public health program funding mechanisms commonly include process evaluation mandates.

Purpose: The use of process mining was explored as a methodology to assess the fidelity of the MIECHV programs’ actual workflow to that of their intended models.

Methods: Research Electronic Data Capture (REDCap) data files that were populated with program process data elements from the local implementing agencies were mined. The focus was on three main variables: participant identification, activity labels, and timestamps. These variables were imported into the Disco process-mining software. Disco was used to develop process maps to track process pathways and compare the actual workflow against the intended model.

Results: Using process mining as a diagnostic tool, fidelity to the MIECHV process model was assessed, identifying a total of 262 different process variations. The 15 most frequent variations represent 60.7% of the total pool of process variations, 13 of which were deemed to have fidelity to the intended model. Analysis of the variations indicated that many activities in the intended process were skipped or implemented out of sequence.

Implications: Process mining is a useful tool for organizations to visually display, track, understand, compare, and improve their workflow processes. This method should be considered by programs as complex as MIECHV to improve the data reporting and the identification of opportunities to strengthen programs.