Sample diagnoses data
Patient diagnoses with ICD-10 codes for healthcare and billing systems
Free to download, no account needed for the 1k-row files below.
Larger sizes — up to 1M rows — need a free account.
This is synthetic data: it looks realistic and refers to nobody, so it
carries none of the data-protection risk of copying production records into a test system.
Downloads
| Size | Format | Locale |
|---|---|---|
| 1k rows | csv | de_DE en_GB en_US es_ES fr_FR ja_JP |
| 1k rows | jsonl | de_DE en_GB en_US es_ES fr_FR ja_JP |
| 1k rows | parquet | de_DE en_GB en_US es_ES fr_FR ja_JP |
| 10k rows | csv | free account (6 locales) |
| 10k rows | jsonl | free account (6 locales) |
| 10k rows | parquet | free account (6 locales) |
| 100k rows | csv | free account (6 locales) |
| 100k rows | jsonl | free account (6 locales) |
| 100k rows | parquet | free account (6 locales) |
| 500k rows | csv | free account (2 locales) |
| 500k rows | jsonl | free account (2 locales) |
| 500k rows | parquet | free account (2 locales) |
| 1M rows | csv | free account (2 locales) |
| 1M rows | jsonl | free account (2 locales) |
Columns
| Column | Type | |
|---|---|---|
id | id | always set |
diagnosis_id | sequence | always set |
patient_id | foreign_key | always set |
encounter_id | foreign_key | always set |
icd10_code | zipfian_choice | always set |
icd10_description | constant | always set |
diagnosis_type | weighted_choice | always set |
present_on_admission | weighted_choice | always set |
diagnosis_priority | distribution_integer | always set |
severity | weighted_choice | always set |
onset_date | date | optional |
resolved_date | date | optional |
is_chronic | weighted_choice | always set |
is_principal | weighted_choice | always set |
diagnosed_by | name | always set |
diagnosis_source | weighted_choice | always set |
status | weighted_choice | always set |
notes | sentence | optional |
created_at | datetime | always set |
updated_at | datetime | always set |
Need a different shape?
Generate this dataset with your own columns, row counts and locales, or browse the full catalogue of 88 ready-made files for this template alone.