Investigating the service utilization and pathways of patients with alcohol use disorders.

BACKGROUND

Process mining was widely used in healthcare settings to analyse patient pathways, providing insights about patient care and resource allocation. This study aimed to apply process mining to explore service utilization and pathways of patients with alcohol use disorders in receiving the community-based Drug and Alcohol Service to establish a baseline for evaluating improvement initiatives.

METHODS

Patients' encounters with the Drug and Alcohol Service and their direct care activities (i.e. Intake, Assessment, and treatment service activities) that occurred in these encounters were analysed to depict the actual pathways. Patterns were uncovered from these actual pathways using a process mining tool Disco. Patients with alcohol use disorders were divided into two groups-patients who use polysubstance and those who do not-to compare their service utilization and pathway patterns.

RESULTS

Five hundred seventy-one patients with alcohol use disorders incurred 1447 encounters, comprising 13,974 activities. The top three actual pathways with at least three activities were: (i) pathway 'Intake-> Withdrawal Management-> Withdrawal Management' (170 (11.7%) encounters and 130 (22.8%) patients); (ii) pathway 'Intake-> Adult Counselling-> Adult Counselling' (161 (11.1%) encounters and 126 (22.1%) patients); and (iii) pathway 'Intake-> Assessment-> Withdrawal Management' (155 (10.7%) encounters and 129 (22.6%) patients). However, 313 (21.6%) encounters did not proceed beyond the Intake stage. When the patients started their pathways from Intake, their immediate next activities were frequently Withdrawal Management, Assessment or Adult Counselling. Compared to those who do not use polysubstance, patients who use polysubstance had a higher proportion of Intake (19.3% vs 11.8%, P < .001), Involuntary Treatment (4.0% vs 1.0%, P < .001) and Magistrate Early Referral into Treatment (6.4% vs .3%, P < .001), but a lower proportion of Withdrawal Management (31.2% vs 39.0%, P < .001) and Adult Counselling (27.8% vs 37.6%, P < .001).

CONCLUSION

This study contributes to an in-depth understanding of service utilization and pathways of patients with alcohol use disorders within the community-based Drug and Alcohol Service. Future studies should investigate their reasons for dropping out after Intake.