Medical Appointment No-Show Records with 19 Features, 2017

Medical-Appointment contains 61,000 patient appointment records from January to April 2017. The dataset includes 19 features related to appointment scheduling, patient demographics, and the target variable indicating whether a patient attended. It is available on OpenML under a GPL 2 license.

Use Cases

• Predict patient no-show probability based on appointment lead time (latencia) and patient demographics.
• Analyze the impact of scheduling channel (canal) and appointment type (tipo) on attendance rates.
• Model seasonal and weekly appointment attendance patterns using derived cyclical date features.
• Identify patient segments with high no-show risk based on age (edad) and specialty (especialidad).

Strengths

• Contains 61,000 data points, providing a substantial sample for analysis.
• Includes 19 features covering scheduling metadata, patient demographics, and engineered cyclical time representations.
• Has a clear binary target variable (show) for supervised learning tasks.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Last update date is unknown; freshness unverified.
• Data may reflect temporal and institutional bias inherent to its source on OpenML.

Provenance

Source

OpenML

Collection Method

Likely collected from a healthcare appointment scheduling system.

Time Range

2017-01-01 to 2017-04-30