Predictive model-based interventions to reduce outpatient no-shows: a rapid systematic review

Area of Science:

• Health Services Research
• Clinical Informatics
• Patient Engagement

Background:

• Outpatient no-shows significantly impact healthcare costs and the quality of patient care.
• Predictive models offer a potential strategy to proactively identify patients at high risk of missing appointments.
• Targeted interventions based on predictive modeling could mitigate the negative consequences of no-shows.

Purpose of the Study:

• To systematically review the effectiveness of interventions based on predictive models for reducing outpatient no-shows.
• To evaluate the impact of these interventions on appointment adherence, costs, patient acceptability, and equity.
• To synthesize evidence from randomized controlled trials (RCTs) and non-RCTs on predictive model-based interventions.

Main Methods:

• A rapid systematic review was conducted, searching multiple databases up to July 2022.
• Included studies were randomized controlled trials (RCTs) and non-randomized studies.
• Data on no-show rates, cost-effectiveness, acceptability, and equity were extracted and analyzed.

Main Results:

• High certainty evidence shows predictive model-based text message reminders reduce no-shows (median RR 0.91).
• Moderate certainty evidence indicates phone call reminders (median RR 0.61) and patient navigators (RR 0.55) are effective.
• The effectiveness of predictive model-based overbooking remains uncertain; limited data exists on cost, acceptability, and equity.

Conclusions:

• Predictive modeling combined with text reminders, phone calls, or patient navigators are likely effective in reducing outpatient no-shows.
• Further research is warranted to compare targeted interventions for high-risk patients against universal interventions.
• The findings support the use of predictive analytics to optimize appointment scheduling and resource allocation.