{
  "fetched_at": "2026-10-19T08:12:44Z",
  "source_url": "https://www.kaggle.com/competitions/eastwood-and-cleef-ml-disease/data",
  "pages": [
    {
      "name": "data-description",
      "content": "This page describes what files have been provided and the format of each. \n\n## Files\n\n*   **admissions_data.csv** - contains information about hospital admissions for 2018-2019\n*   **disease_data.csv** - contains information about ICD and HCC for 2018-2019\n*   **labs_data.csv** - contains information about laboratory tests for 2018-2019\n*   **patients_data.csv** - contains member data such as gender, age, etc. for 2018-2019\n*   **prescriptions_data.csv** - contains drug prescriptions for 2018-2019\n*   **train_labels.csv** - contains labels for 2019 (new diseases discovered that year)\n*   **sample_submission.csv** - your prediction with HCCs for 2020\n\n## Columns\n\n*   `member_id` - patients' ID\n*   `admission_date` - a date when hospital admission occurred\n*   `discharge_date` -  date when member was discharge from the hospital\n*   `admission_type` - a hospital admission type\n*   `icd_code` - an ICD code (diagnosis) used for hospital admission\n*   `is_readmission` - 1 if member was already in the hospital in last 30 days, otherwise 0\n*   `er_to_inp_admission_transfer` - 1 if member had inpatient (INP) and ER (emergency) hospital admission same day, otherwise 0\n*   `days_to_prev_admission` - a number of days since last admission\n*   `year_of_service` - a year when ICD / HCC was detected by doctors\n*   `icd_code` - an ICD code (disease)\n*   `icd_chronic_or_acute` - a chronic / acute disease type\n*   `hcc_code` - a HCC coding for ICD\n*   `hcc_chronic_or_acute` - a chronic / acute HCC type\n*   `date_of_service` - a date when some service was provided to patients\n*   `loinc_code` - a laboratory test code\n*   `performed_test_name ` - a laboratory test name\n*   `abnormal_code` - a flag that tells about low (L) or high (H) lab results\n*   `result_value` - a result value\n*   `normal_low_value_numeric` - a lower bound for reference lab result\n*   `normal_high_value_numeric` - an upper bound for reference lab result\n*   `date_of_service` - a date when some service was provided to patients\n*   `patient_age` - a patient's age\n*   `patient_gender` - a patient's gender\n*   `patient_age` - a patient's age\n*   `dual_status` - a flag that tells about Medicaid / Medicate insurance programms\n*   `patient_age` - a patient's age\n*   `insurance_company` - a company providing medical insurance\n*   `insurance_type` - an insurance type\n*   `pbp_type` - a member's risk group\n*   `county` - a member's county\n*   `city` - a member's city\n*   `date_filled` - a date when prescription was filled for the member\n*   `ndc_number` - an unique drug ID \n*   `days_supply` - a number of days for which the prescription was filled\n*   `drug_name` - a drug name\n*   `metric_quantity` - a drug dosage\n*   `new_or_refill` - 1 if a new drug was prescribed, otherwise 0\n*   `hcc_9` - a flag that identifies that member has HCC 9. Similar logic should be used for other hcc_N columns\n\n\n\n\n\n\n\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "You need to predict new diseases for patients.\n\n`Train data` contains medical information for 2018-2019 period and labels with new diseases discovered in 2019.\n\nYou are required to predict `new diseases for 2020` using data for 2018-2019.\n\n## Background\n\nHierarchical condition category relies on ICD-10 coding to assign risk scores to patients. Each HCC is mapped to an ICD-10 code. Along with demographic factors (such as age and gender), insurance companies use HCC coding to assign patients a risk adjustment factor. It is very important to predict HCC coding to evaluate health risk of patients.\n\nEvery HCC can be mapped to several ICDs but not every ICD has a corresponding HCC. It is a one to many connection for HCC to ICDs.\n\nWe are providing ICD history of patients to predict their net new HCC coding for next year.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Submissions are evaluated on the `mean column-wise ROC AUC`. In other words, the score is the average of the individual AUCs of each predicted column.\n\n## Submission Format\n\n**For every patient in the dataset**, submission files should contain two columns: `member_id` and columns with diseases: `hcc_9`,`hcc_10`,`hcc_11`,`hcc_12`,`hcc_18`,`hcc_19`,`hcc_21`,`hcc_22`,`hcc_23`,`hcc_40`,`hcc_47`,`hcc_48`,`hcc_51`,`hcc_52`,`hcc_55`,`hcc_59`,`hcc_75`,`hcc_79`,`hcc_84`,`hcc_85`,`hcc_88`,`hcc_96`,`hcc_103`,`hcc_108`,`hcc_111`,`hcc_112`,`hcc_124`,`hcc_137`,`hcc_138`,`hcc_161`. \n\nThe file should contain a header and have the following format:\n```\nmember_id,hcc_9,hcc_10,hcc_11,hcc_12,hcc_18,hcc_19,hcc_21,hcc_22,hcc_23,hcc_40,hcc_47,hcc_48,hcc_51,hcc_52,hcc_55,hcc_59,hcc_75,hcc_79,hcc_84,hcc_85,hcc_88,hcc_96,hcc_103,hcc_108,hcc_111,hcc_112,hcc_124,hcc_137,hcc_138,hcc_161\n\n1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0\n```",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "- Only single participants allowed\n- Data can be used only for the purpose of this competition",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}