Type-1 Diabetes Code Screening in Patient Records

Easy

A clinical data platform stores patient diagnosis tags as a space-separated code list. Analysts need a quick filter that identifies records with Type-1 diabetes-related codes. Table: patients patient_id: unique patient identifier patient_name: patient display name conditions: optional space-separated diagnosis code string Matching rule: A patient qualifies when at least one diagnosis token starts with the prefix DIAB1. Matching is token-based, not arbitrary substring matching inside a token. Task: Return all qualifying patients. Output requirements: Return exactly these columns in this order: patient_id patient_name conditions Output row order can be arbitrary. Supported submission environments: SQL (SQLite 3.27.2) Pandas (Python)

Example 1

Input

patients:
| patient_id | patient_name | conditions      |
|------------|--------------|-----------------|
| 1          | Daniel       | YFEV COUGH      |
| 2          | Alice        |                 |
| 3          | Bob          | DIAB100 MYOP    |
| 4          | George       | ACNE DIAB100    |
| 5          | Alain        | DIAB201         |

Output

[
  {"patient_id":3,"patient_name":"Bob","conditions":"DIAB100 MYOP"},
  {"patient_id":4,"patient_name":"George","conditions":"ACNE DIAB100"}
]

Explanation

Patients 3 and 4 each contain a diagnosis token that starts with DIAB1. Patient 5 has DIAB201, which does not match the required prefix.

Example 2

Input

patients:
| patient_id | patient_name | conditions         |
|------------|--------------|--------------------|
| 10         | Iris         | XDIAB100 ASTH01    |
| 11         | Noah         | PREDIAB100 COUGH   |
| 12         | Riya         | ASTH01 FLU11       |

Output

[]

Explanation

No diagnosis token starts with DIAB1. Tokens that only contain DIAB1 as an internal substring are not valid matches.

Example 3

Input

patients:
| patient_id | patient_name | conditions                |
|------------|--------------|---------------------------|
| 20         | Mina         | DIAB10                    |
| 21         | Owen         | HBP22 DIAB1X RESP05       |
| 22         | Pia          | NULL                      |
| 23         | Quinn        |                           |

Output

[
  {"patient_id":20,"patient_name":"Mina","conditions":"DIAB10"},
  {"patient_id":21,"patient_name":"Owen","conditions":"HBP22 DIAB1X RESP05"}
]

Explanation

Both DIAB10 and DIAB1X begin with DIAB1, so those rows qualify. Null and empty condition strings do not qualify.

Constraints

0 <= number of rows in patients <= 300000
patient_id is unique in patients
1 <= patient_id <= 1000000000
patient_name is a non-empty string
conditions may be NULL, empty, or a space-separated list of diagnosis codes
Diagnosis tokens are matched by prefix DIAB1
Return exactly three columns: patient_id, patient_name, conditions
Output row order can be arbitrary
For SQL submissions, use SQLite 3.27.2 semantics
For Pandas submissions, return a DataFrame with columns: patient_id, patient_name, conditions

patients: | patient_id | patient_name | conditions | |------------|--------------|-----------------| | 1 | Daniel | YFEV COUGH | | 2 | Alice | | | 3 | Bob | DIAB100 MYOP | | 4 | George | ACNE DIAB100 | | 5 | Alain | DIAB201 |

patients: | patient_id | patient_name | conditions | |------------|--------------|--------------------| | 10 | Iris | XDIAB100 ASTH01 | | 11 | Noah | PREDIAB100 COUGH | | 12 | Riya | ASTH01 FLU11 |

patients: | patient_id | patient_name | conditions | |------------|--------------|---------------------------| | 20 | Mina | DIAB10 | | 21 | Owen | HBP22 DIAB1X RESP05 | | 22 | Pia | NULL | | 23 | Quinn | |