Investigating EHR data from PIC-SURE - python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f15c7b43",
   "metadata": {},
   "source": [
    "# Investigating EHR data from PIC-SURE"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10c76af6",
   "metadata": {},
   "source": [
    "The purpose of this notebook is to help researchers get started with EHR analysis in python using clinical data exported from PIC-SURE. \n",
    "\n",
    "### About the data\n",
    "This example code uses a synthetically generated dataset from Synthea, an open-source patient population simulation. \n",
    "The dataset has been formatted to mirror what a data export from PIC-SURE may look like.  \n",
    "It includes all age groups and is not limited to pediatric patients. \n",
    "\n",
    "\n",
    "\n",
    "Jason Walonoski, Mark Kramer, Joseph Nichols, Andre Quina, Chris Moesel, Dylan Hall, Carlton Duffett, Kudakwashe Dube, Thomas Gallagher, Scott McLachlan, Synthea: An approach, method, and software mechanism for generating synthetic patients and the synthetic electronic health care record, Journal of the American Medical Informatics Association, Volume 25, Issue 3, March 2018, Pages 230–238, https://doi.org/10.1093/jamia/ocx079\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80c82c23",
   "metadata": {},
   "source": [
    "###  Installing packages\n",
    "The following code installs and loads packages needed for basic data management and manipulation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "640d6363",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/smakwana/Library/Python/3.9/lib/python/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from datetime import datetime\n",
    "import requests\n",
    "import zipfile\n",
    "import io\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3d8dfa59",
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_colwidth', None)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "baadad31",
   "metadata": {},
   "source": [
    "## Loading the data\n",
    "The following code downloads the dataset. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e968845e",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files in ZIP: ['synthea_10k_picsure_format.csv', '__MACOSX/._synthea_10k_picsure_format.csv']\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/nz/k_g_v4910fd_wthf2ngtf13c0000gq/T/ipykernel_38006/1523992457.py:15: DtypeWarning: Columns (3) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  data = pd.read_csv(f)\n"
     ]
    }
   ],
   "source": [
    "# The data is stored on github:\n",
    "url = \"https://github.com/hms-dbmi/pic-sure-public-datasets/raw/main/synthea_10k_picsure_format.csv.zip\"\n",
    "\n",
    "# Download the ZIP file from GitHub\n",
    "response = requests.get(url)\n",
    "response.raise_for_status()  # Ensure the request was successful\n",
    "\n",
    "# Open the ZIP file\n",
    "with zipfile.ZipFile(io.BytesIO(response.content)) as z:\n",
    "    # List files in the ZIP\n",
    "    print(\"Files in ZIP:\", z.namelist())\n",
    "    \n",
    "    # Extract and load the CSV file (assuming the first file is the desired CSV)\n",
    "    with z.open(z.namelist()[0]) as f:\n",
    "        data = pd.read_csv(f)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0be610ef",
   "metadata": {},
   "source": [
    "## Taking a first look at the dataset\n",
    "Let's use the `head()` function to take a first look at the dataset. \n",
    "You'll see 5 columns:\n",
    "- PATIENT_NUM: Represents a unique patient identifier\n",
    "- CONCEPT_PATH: Represents the clinical variable for a given observation (row)\n",
    "- TVAL_CHAR: The text / character value of the clinical variable, if applicable\n",
    "- NVAL_NUM: The numerical value of the clinical variable, if applicable\n",
    "- TIMESTAMP: The date and time associated to a given observation (row)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "22353812",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>CONCEPT_PATH</th>\n",
       "      <th>NVAL_NUM</th>\n",
       "      <th>TVAL_CHAR</th>\n",
       "      <th>TIMESTAMP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3737898</th>\n",
       "      <td>11642</td>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Radiologic examination, chest</td>\n",
       "      <td>2015-08-27 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3737899</th>\n",
       "      <td>11642</td>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Radiologic examination</td>\n",
       "      <td>2015-08-27 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3737900</th>\n",
       "      <td>11643</td>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Radiologic examination, chest</td>\n",
       "      <td>2017-04-23 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3737901</th>\n",
       "      <td>11643</td>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Radiologic examination</td>\n",
       "      <td>2017-04-23 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3737902</th>\n",
       "      <td>11642</td>\n",
       "      <td>\\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Urinary System\\Surgical Procedures on the Kidney</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Renal Transplantation Procedures</td>\n",
       "      <td>2018-03-24 04:00:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         PATIENT_NUM  \\\n",
       "3737898        11642   \n",
       "3737899        11642   \n",
       "3737900        11643   \n",
       "3737901        11643   \n",
       "3737902        11642   \n",
       "\n",
       "                                                                                                                                                            CONCEPT_PATH  \\\n",
       "3737898  \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3737899  \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3737900  \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3737901  \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3737902                                                          \\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Urinary System\\Surgical Procedures on the Kidney   \n",
       "\n",
       "         NVAL_NUM                         TVAL_CHAR           TIMESTAMP  \n",
       "3737898       NaN     Radiologic examination, chest 2015-08-27 04:00:00  \n",
       "3737899       NaN            Radiologic examination 2015-08-27 04:00:00  \n",
       "3737900       NaN     Radiologic examination, chest 2017-04-23 04:00:00  \n",
       "3737901       NaN            Radiologic examination 2017-04-23 04:00:00  \n",
       "3737902       NaN  Renal Transplantation Procedures 2018-03-24 04:00:00  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The timestamp is in unix format. Convert to datetime\n",
    "data['TIMESTAMP'] = pd.to_datetime(data['TIMESTAMP'], unit='s')\n",
    "\n",
    "data.tail()#"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d0208c8",
   "metadata": {},
   "source": [
    "The EHR data represented in this dataset may be very large, making it difficult to explore manually. \n",
    "\n",
    "You'll notice that the `CONCEPT_PATH` column consists of a hierarchy that narrows down to a certain clinical variable. \n",
    "\n",
    "Let's take a look at the highest level of the `CONCEPT_PATH` to see what kind of data we have. We can consider this to be the `VARIABLE_TYPE`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d9dd50c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "VARIABLE_TYPE\n",
       "ACT Lab Test Results    2788447\n",
       "ACT Medications          748721\n",
       "ACT Diagnosis ICD-10      96705\n",
       "ACT Procedures CPT        55808\n",
       "ACT Demographics          48222\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Isolate the variable type using regular expressions, and create a new column for variable type\n",
    "data['VARIABLE_TYPE'] = data['CONCEPT_PATH'].str.extract(r'\\\\([^\\\\]+)\\\\')\n",
    "# Count how many observations are associated with each variable type\n",
    "data['VARIABLE_TYPE'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08aa83b7",
   "metadata": {},
   "source": [
    "As seen above, this dataset contains the following data types: \n",
    "- ACT Lab Test Results (Labs)\n",
    "- ACT Medications\n",
    "- ACT Diagnosis ICD-10 (Diagnoses)\n",
    "- ACT Procedures CPT (Procedures)\n",
    "- ACT Demographics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a123bda5",
   "metadata": {},
   "source": [
    "This dataset should contain about 10,000 synthetic patient records. Let's verify this by checking the count of unique patient IDs present in the dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "03729a95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11644"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['PATIENT_NUM'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b749adfe",
   "metadata": {},
   "source": [
    "Let's also check what date ranges are represented in our dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "df600579",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Changing the time stamp to a date format\n",
    "data['TIMESTAMP'] = pd.to_datetime(data['TIMESTAMP'], errors='coerce')\n",
    "data['DATE'] = data['TIMESTAMP'].dt.date\n",
    "\n",
    "\n",
    "# Create a histogram of dates\n",
    "plt.figure(figsize=(12, 6))\n",
    "data['DATE'].hist(bins=30, color='skyblue', edgecolor='black')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Frequency')\n",
    "plt.title('Histogram of Dates')\n",
    "plt.xticks(rotation=45)\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb2fd7d9",
   "metadata": {},
   "source": [
    "The majority of our dataset is from the last 15 years, as expected. \n",
    "\n",
    "We also see data stretching far back to 1920; perhaps this is associated with a patient's age. Depending on your research question, you may want to investigate those outliers. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef552dd9",
   "metadata": {},
   "source": [
    "## Exploring Demographic Data\n",
    "Here we will explore, organize and visualize the demographic data available in our dataset. This includes age, sex and race.\n",
    "\n",
    "Let's create a new dataframe called `demo_df` that will only include the demographics information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "9ce1c34e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "demo_df = data[data['VARIABLE_TYPE'] == 'ACT Demographics']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59720149",
   "metadata": {},
   "source": [
    "Let's first take a look at the **sex** breakdown. \n",
    "\n",
    "We can find the Sex related observations by searching for this value in the CONCEPT_PATH column. Then, we take the sex related values (found in the TVAL_CHAR column) and transform them into a new column, \"sex\". \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "fb986f34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sex</th>\n",
       "      <th>count</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Female</td>\n",
       "      <td>5747</td>\n",
       "      <td>49.355891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Male</td>\n",
       "      <td>5897</td>\n",
       "      <td>50.644109</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      sex  count  percentage\n",
       "0  Female   5747   49.355891\n",
       "1    Male   5897   50.644109"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#isolating and showing sex data:\n",
    "sex_df = demo_df[demo_df['CONCEPT_PATH'].str.contains(r'\\\\Sex\\\\', na=False)]\n",
    "\n",
    "#creating a new dataset only including the patient id#, sex and changing column name to 'sex':\n",
    "sex_df = sex_df[['PATIENT_NUM', 'TVAL_CHAR']].rename(columns={'TVAL_CHAR': 'sex'})\n",
    "\n",
    "sex_df \n",
    "\n",
    "\n",
    "#grouping by sex and calculating the percentage of each:\n",
    "sex_count = (sex_df.groupby('sex')\n",
    "              .size()\n",
    "              .reset_index(name='count'))\n",
    "sex_count['percentage'] = (sex_count['count'] / sex_count['count'].sum()) * 100\n",
    "\n",
    "sex_count\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17d3a9a6",
   "metadata": {},
   "source": [
    "Now let's look at the **age** distribution of our patients. \n",
    "\n",
    "Below, we isolate the age data by searching the CONCEPT_PATH for Age. Then, we take the age related values (found in the NVAL_NUM column) and transform them into a new column, \"age\". "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "94dd1187",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36578</th>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36579</th>\n",
       "      <td>2</td>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36580</th>\n",
       "      <td>3</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36581</th>\n",
       "      <td>4</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36582</th>\n",
       "      <td>5</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48217</th>\n",
       "      <td>11640</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48218</th>\n",
       "      <td>11641</td>\n",
       "      <td>74.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48219</th>\n",
       "      <td>11642</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48220</th>\n",
       "      <td>11643</td>\n",
       "      <td>77.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48221</th>\n",
       "      <td>11644</td>\n",
       "      <td>70.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11644 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       PATIENT_NUM   age\n",
       "36578            1   5.0\n",
       "36579            2  15.0\n",
       "36580            3   8.0\n",
       "36581            4  60.0\n",
       "36582            5   9.0\n",
       "...            ...   ...\n",
       "48217        11640  57.0\n",
       "48218        11641  74.0\n",
       "48219        11642  65.0\n",
       "48220        11643  77.0\n",
       "48221        11644  70.0\n",
       "\n",
       "[11644 rows x 2 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# isolating and showing age data:\n",
    "age_df = demo_df[demo_df['CONCEPT_PATH'].str.contains(r'\\\\Age\\\\')]\n",
    "\n",
    "#creating a new dataset only including the patient id#, age and changing column name to 'age':\n",
    "age_df = age_df[['PATIENT_NUM', 'NVAL_NUM']].rename(columns={'NVAL_NUM': 'age'})\n",
    "\n",
    "age_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff2f61d6",
   "metadata": {},
   "source": [
    "We can now use this age dataframe to plot the age distribution in our data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "21c55f00",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Creating a histogram demonstrating the age distribution:\n",
    "\n",
    "plt.figure(figsize=(8, 5))\n",
    "sns.histplot(age_df['age'], bins=110, color='blue', edgecolor='black')\n",
    "plt.title('Age Distribution')\n",
    "plt.xlabel('Age')\n",
    "plt.ylabel('Frequency')\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71cd65a7",
   "metadata": {},
   "source": [
    "We can also group our patients into age groups and create a table detailing how many patients are in each age group. \n",
    "\n",
    "We will group our patients into the following groups: \n",
    "- under 18\n",
    "- 18-65\n",
    "- 65+"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "91ffe9fb",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age_group</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>&lt;18</td>\n",
       "      <td>2144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18-65</td>\n",
       "      <td>7437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>65+</td>\n",
       "      <td>2063</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  age_group  count\n",
       "0       <18   2144\n",
       "1     18-65   7437\n",
       "2       65+   2063"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# grouping by age (<18 , 18-65, 65+)\n",
    "\n",
    "age_group= age_df.copy()\n",
    "\n",
    "bins = [-float('inf'), 17, 64, float('inf')]\n",
    "labels = ['<18', '18-65', '65+']\n",
    "\n",
    "age_group['age_group'] = pd.cut(age_group['age'], bins=bins, labels=labels, right=True)\n",
    "\n",
    "\n",
    "# Calculate the number of patients in each age group\n",
    "age_group_sum = age_group['age_group'].value_counts().reindex(labels).reset_index()\n",
    "age_group_sum.columns = ['age_group', 'count']\n",
    "\n",
    "age_group_sum"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8fdf0e84",
   "metadata": {},
   "source": [
    "Finally, let's investigate the **race** distribution. \n",
    "\n",
    "Similarly to above, we isolate the race data by searching the CONCEPT_PATH for Race. Then, we take the race related values (found in the TVAL_CHAR column) and transform them into a new column, \"race\". "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "24f9b5cf",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#isolating and showing race data:\n",
    "race_data = demo_df[demo_df['CONCEPT_PATH'].str.contains(r'\\\\Race\\\\', na=False)]\n",
    "\n",
    "#creating a new dataset only including the patient id#, age and changing column name to 'age':\n",
    "race_data = race_data[['PATIENT_NUM', 'TVAL_CHAR']].rename(columns={'TVAL_CHAR': 'race'})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "887634ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>percentage</th>\n",
       "      <th>race</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>race</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>White</th>\n",
       "      <td>9648</td>\n",
       "      <td>82.858124</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Black or African American</th>\n",
       "      <td>918</td>\n",
       "      <td>7.883889</td>\n",
       "      <td>Black or African American</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Asian</th>\n",
       "      <td>786</td>\n",
       "      <td>6.750258</td>\n",
       "      <td>Asian</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Native Hawaiian or Other Pacific Islander</th>\n",
       "      <td>118</td>\n",
       "      <td>1.013397</td>\n",
       "      <td>Native Hawaiian or Other Pacific Islander</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>No Information</th>\n",
       "      <td>113</td>\n",
       "      <td>0.970457</td>\n",
       "      <td>No Information</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Indian or Alaska Native</th>\n",
       "      <td>61</td>\n",
       "      <td>0.523875</td>\n",
       "      <td>American Indian or Alaska Native</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           count  percentage  \\\n",
       "race                                                           \n",
       "White                                       9648   82.858124   \n",
       "Black or African American                    918    7.883889   \n",
       "Asian                                        786    6.750258   \n",
       "Native Hawaiian or Other Pacific Islander    118    1.013397   \n",
       "No Information                               113    0.970457   \n",
       "American Indian or Alaska Native              61    0.523875   \n",
       "\n",
       "                                                                                race  \n",
       "race                                                                                  \n",
       "White                                                                          White  \n",
       "Black or African American                                  Black or African American  \n",
       "Asian                                                                          Asian  \n",
       "Native Hawaiian or Other Pacific Islander  Native Hawaiian or Other Pacific Islander  \n",
       "No Information                                                        No Information  \n",
       "American Indian or Alaska Native                    American Indian or Alaska Native  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#grouping by race and calculating the percentage of each:\n",
    "\n",
    "count= race_data['race'].value_counts()\n",
    "percentages = race_data['race'].value_counts(normalize=True) * 100\n",
    "\n",
    "race_count= pd.DataFrame({\n",
    "    'count': count,\n",
    "    'percentage': percentages, \n",
    "    'race' : count.index\n",
    "    })\n",
    "race_count"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff5e93c0",
   "metadata": {},
   "source": [
    "We will plot these results in a bar plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "1b256ecf",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/nz/k_g_v4910fd_wthf2ngtf13c0000gq/T/ipykernel_38006/1025067948.py:3: FutureWarning: \n",
      "\n",
      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
      "\n",
      "  sns.barplot(data=race_count, x='race', y='percentage', palette='viridis')\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#create a bar plot representing race distribution:\n",
    "plt.figure(figsize=(10, 6))\n",
    "sns.barplot(data=race_count, x='race', y='percentage', palette='viridis')\n",
    "plt.title('Percentage by Race')\n",
    "plt.xlabel('Race')\n",
    "plt.ylabel('Percentage')\n",
    "plt.xticks(rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "119f6c8a",
   "metadata": {},
   "source": [
    "The way our initial dataset is formatted can be difficult to view multiple variables about a given patient at the same time. \n",
    "\n",
    "Below, we create one table presenting all demographic data (age, sex, race) by patient ID. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "e5df8862",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>race</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>8.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>60.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11639</th>\n",
       "      <td>11640</td>\n",
       "      <td>57.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11640</th>\n",
       "      <td>11641</td>\n",
       "      <td>74.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11641</th>\n",
       "      <td>11642</td>\n",
       "      <td>65.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>Black or African American</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11642</th>\n",
       "      <td>11643</td>\n",
       "      <td>77.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11643</th>\n",
       "      <td>11644</td>\n",
       "      <td>70.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>Black or African American</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11644 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       PATIENT_NUM   age     sex                       race\n",
       "0                1   5.0  Female                      White\n",
       "1                2  15.0    Male                      White\n",
       "2                3   8.0    Male                      White\n",
       "3                4  60.0  Female                      White\n",
       "4                5   9.0  Female                      White\n",
       "...            ...   ...     ...                        ...\n",
       "11639        11640  57.0    Male                      White\n",
       "11640        11641  74.0  Female                      White\n",
       "11641        11642  65.0  Female  Black or African American\n",
       "11642        11643  77.0    Male                      White\n",
       "11643        11644  70.0  Female  Black or African American\n",
       "\n",
       "[11644 rows x 4 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merging all columns to show age, sex, race in one table:\n",
    "\n",
    "first_merge = pd.merge(age_df[['PATIENT_NUM', 'age']], sex_df[['PATIENT_NUM', 'sex']], on='PATIENT_NUM', how='left')\n",
    "\n",
    "demographics_joined = pd.merge(first_merge, race_data[['PATIENT_NUM', 'race']], on='PATIENT_NUM', how='left')\n",
    "\n",
    "demographics_joined"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53546716",
   "metadata": {},
   "source": [
    "## Investigating ICD codes: Epilepsy case study"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "123e6fec",
   "metadata": {},
   "source": [
    "We can isolate the diagnosis related information by searching for 'ACT Diagnosis' in the CONCEPT_PATH column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "38614c81",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#create a new data frame including only the patients diagnosis data:\n",
    "\n",
    "diagnosis_icd10 = data[data['CONCEPT_PATH'].str.contains(r'\\\\ACT Diagnosis ICD-10\\\\')]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4734862",
   "metadata": {},
   "source": [
    "### In this case study, we will investigate epilepsy as the disease of interest. \n",
    "\n",
    "Epilapsy is represented by  ICD group code \"G40\", which includes the following and their sub-groups:\n",
    "- G40.0: Localization-related (focal) (partial) idiopathic epilepsy and epileptic syndromes with seizures of localized onset.\n",
    "- G40.1: Localization-related (focal) (partial) symptomatic epilepsy and epileptic syndromes with simple partial seizures.\n",
    "- G40.2: Localization-related (focal) (partial) symptomatic epilepsy and epileptic syndromes with complex partial seizures.\n",
    "- G40.3: Generalized idiopathic epilepsy and epileptic syndromes\n",
    "- G40.A: Absence epileptic syndrome.\n",
    "- G40.B: Juvenile myoclonic epilepsy [impulsive petit mal].\n",
    "- G40.C: Lafora progressive myoclonus epilepsy, intractable.\n",
    "- G40.4: Other generalized epilepsy and epileptic syndromes.\n",
    "- G40.5: Epileptic seizures related to external causes.\n",
    "- G40.8: Other epilepsy and recurrent seizures.\n",
    "- G40.9: Epilepsy, unspecified.\n",
    "\n",
    "*A full list of ICD-10 codes can be found at https://www.icd10data.com/ICD10CM/Codes*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cfa41de",
   "metadata": {},
   "source": [
    "### Isolating the patient cohort\n",
    "\n",
    "First, we create a new data-frame that only includes patients with epilepsy.\n",
    "\n",
    "We create a variable called `icd_code_of_interest` and set it to `G40`.\n",
    "\n",
    "(you can replace the ICD code \"G40\" with the ICD code of your choice if you are interested in investigating a different disease group!):\n",
    "\n",
    "You can see a preview of this epilepsy focused dataset below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d49bab50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>CONCEPT_PATH</th>\n",
       "      <th>icd_10_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>48255</th>\n",
       "      <td>9</td>\n",
       "      <td>\\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable</td>\n",
       "      <td>G40.909 Epilepsy, unspecified, not intractable, without status epilepticus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48256</th>\n",
       "      <td>9</td>\n",
       "      <td>\\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable</td>\n",
       "      <td>G40.909 Epilepsy, unspecified, not intractable, without status epilepticus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48257</th>\n",
       "      <td>9</td>\n",
       "      <td>\\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable</td>\n",
       "      <td>G40.909 Epilepsy, unspecified, not intractable, without status epilepticus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48258</th>\n",
       "      <td>9</td>\n",
       "      <td>\\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable</td>\n",
       "      <td>G40.909 Epilepsy, unspecified, not intractable, without status epilepticus</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48259</th>\n",
       "      <td>9</td>\n",
       "      <td>\\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable</td>\n",
       "      <td>G40.909 Epilepsy, unspecified, not intractable, without status epilepticus</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PATIENT_NUM  \\\n",
       "48255            9   \n",
       "48256            9   \n",
       "48257            9   \n",
       "48258            9   \n",
       "48259            9   \n",
       "\n",
       "                                                                                                                                                                                                                                   CONCEPT_PATH  \\\n",
       "48255  \\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable   \n",
       "48256  \\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable   \n",
       "48257  \\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable   \n",
       "48258  \\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable   \n",
       "48259  \\ACT Diagnosis ICD-10\\G00-G99 Diseases of the nervous system (G00-G99)\\G40-G47 Episodic and paroxysmal disorders (G40-G47)\\G40 Epilepsy and recurrent seizures\\G40.9 Epilepsy, unspecified\\G40.90 Epilepsy, unspecified, not intractable   \n",
       "\n",
       "                                                                      icd_10_code  \n",
       "48255  G40.909 Epilepsy, unspecified, not intractable, without status epilepticus  \n",
       "48256  G40.909 Epilepsy, unspecified, not intractable, without status epilepticus  \n",
       "48257  G40.909 Epilepsy, unspecified, not intractable, without status epilepticus  \n",
       "48258  G40.909 Epilepsy, unspecified, not intractable, without status epilepticus  \n",
       "48259  G40.909 Epilepsy, unspecified, not intractable, without status epilepticus  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd_code_of_interest = 'G40'\n",
    "\n",
    "#filtering to include only patients diagnosed with epilepsy and recurrent seizures (ICD GROUP CODE G40):\n",
    "\n",
    "epilepsy_icd10 = diagnosis_icd10[diagnosis_icd10['TVAL_CHAR'].str.contains(icd_code_of_interest)]\n",
    "epilepsy_icd10 = epilepsy_icd10[['PATIENT_NUM', 'CONCEPT_PATH', 'TVAL_CHAR']].rename(columns={'TVAL_CHAR': 'icd_10_code'})\n",
    "\n",
    "# creating a final list of patient_ids with epilepsy (G40 group code), removing duplicates entries for same patient id :\n",
    "epilepsy_icd10_patient_ids = pd.DataFrame(epilepsy_icd10['PATIENT_NUM'].unique(), columns=['PATIENT_NUM'])\n",
    "\n",
    "epilepsy_icd10.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14740839",
   "metadata": {},
   "source": [
    "Let's also isolate the demographic data for patients who have been diagnosed with epilepsy, and see how much of our dataset they represent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "36785e1d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.59 % of patients in our dataset have been diagnosed with an epilepsy ICD10 code.\n"
     ]
    }
   ],
   "source": [
    "# merging the patient id list with the demographic data\n",
    "\n",
    "epi_demo= pd.merge(epilepsy_icd10_patient_ids, demographics_joined, on='PATIENT_NUM', how='left')\n",
    "\n",
    "# calculating percentage of patients with epilepsy from total dataset cohort:\n",
    "\n",
    "epilepsy_percentage = (len(epilepsy_icd10_patient_ids) / data['PATIENT_NUM'].nunique()) * 100           \n",
    "print(round(epilepsy_percentage, 2), \"% of patients in our dataset have been diagnosed with an epilepsy ICD10 code.\")  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2ebbc53",
   "metadata": {},
   "source": [
    "### Investigating the presence of other diagnostic codes in our defined patient cohort\n",
    "Our cohort consists of patients with epilepsy. \n",
    "Let's say we are interested in comparing which patients with epilepsy also have been diagnosed with migranes (ICD GROUP CODE G43).\n",
    "\n",
    "We create a variable called `icd` and set it to `G43`. (you can replace the ICD code \"G43\" with the ICD code of your choice in the next chunk of code)\n",
    "\n",
    "We then search for this ICD code of interest within our cohort. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "f4212407",
   "metadata": {},
   "outputs": [],
   "source": [
    "icd = 'G43'\n",
    "# first, creating a dataframe including only patients diagnosed with migraines (ICD GROUP CODE G43)\n",
    "\n",
    "migraines_icd10 = (diagnosis_icd10[diagnosis_icd10['TVAL_CHAR'].str.contains(icd)]\n",
    "                  .loc[:, ['PATIENT_NUM', 'CONCEPT_PATH', 'TVAL_CHAR']]\n",
    "                  .rename(columns={'TVAL_CHAR': 'icd_10_code'}))\n",
    "\n",
    "#next, extracting only the patient id's and removing any duplicates.\n",
    "migraines_icd10_patient_ids = pd.DataFrame(migraines_icd10['PATIENT_NUM'].unique(), columns=['PATIENT_NUM'])\n",
    "\n",
    "#finally, merging the epilepsy patient id list with the migraine patient id list to create a list including only patients that have both epilepsy and migraine ICD10 codes \n",
    "\n",
    "epilepsy_migraine= pd.merge(epilepsy_icd10_patient_ids, migraines_icd10_patient_ids, on='PATIENT_NUM', how='inner')\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96d15fef",
   "metadata": {},
   "source": [
    "Now, let's examine the original epilepsy dataframe and add a column indicating which patients also have migraines:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b957d7e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>migraines</th>\n",
       "      <th>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>race</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>Black or African American</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>87</td>\n",
       "      <td>0</td>\n",
       "      <td>62.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>122</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>Asian</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>149</td>\n",
       "      <td>0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>175</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>413</th>\n",
       "      <td>11487</td>\n",
       "      <td>0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>414</th>\n",
       "      <td>11504</td>\n",
       "      <td>0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>415</th>\n",
       "      <td>11527</td>\n",
       "      <td>0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>Female</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>416</th>\n",
       "      <td>11546</td>\n",
       "      <td>0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>417</th>\n",
       "      <td>11630</td>\n",
       "      <td>0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>Male</td>\n",
       "      <td>White</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>418 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PATIENT_NUM  migraines   age     sex                       race\n",
       "0              9          0  41.0    Male  Black or African American\n",
       "1             87          0  62.0    Male                      White\n",
       "2            122          0   1.0  Female                      Asian\n",
       "3            149          0  15.0    Male                      White\n",
       "4            175          0   4.0    Male                      White\n",
       "..           ...        ...   ...     ...                        ...\n",
       "413        11487          0  44.0    Male                      White\n",
       "414        11504          0  68.0    Male                      White\n",
       "415        11527          0  29.0  Female                      White\n",
       "416        11546          0  46.0    Male                      White\n",
       "417        11630          0  71.0    Male                      White\n",
       "\n",
       "[418 rows x 5 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# merging the joint list to the original epilepsy patient list, and creating a migraine column indicating if the patient with epilepsy also has a migraine diagnosis (yes=1/ no=0):\n",
    "\n",
    "epilepsy_migraine['migraines'] = 1\n",
    "\n",
    "epilepsy_merged = (pd.merge(epilepsy_icd10_patient_ids, epilepsy_migraine, on='PATIENT_NUM', how='left')\n",
    "                   .fillna({'migraines': 0})\n",
    "                   .astype({'migraines': int})\n",
    "                   .merge(epi_demo, on='PATIENT_NUM', how='left')\n",
    "                   .sort_values(by='PATIENT_NUM'))\n",
    "\n",
    "epilepsy_merged"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c647d55",
   "metadata": {},
   "source": [
    "The following table shows the percentage of patients with migraines in the epilepsy patient cohort.\n",
    "\n",
    "We can see that only a small percentage of our patients with epilepsy also experience migraines."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "45e6e1cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>migraines</th>\n",
       "      <th>count</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>398</td>\n",
       "      <td>95.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>4.78</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   migraines  count  percentage\n",
       "0          0    398       95.22\n",
       "1          1     20        4.78"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#calculating the number and percentage of patient with migraines in the epilepsy patient cohort\n",
    "\n",
    "epi_mig_count = epilepsy_merged.groupby('migraines').size().reset_index(name='count')\n",
    "epi_mig_count['percentage'] = round((epi_mig_count['count'] / epi_mig_count['count'].sum()) * 100, 2)\n",
    "\n",
    "epi_mig_count\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ad7ab77",
   "metadata": {},
   "source": [
    "### Investigating what medications have been administered to our cohort of interest\n",
    "\n",
    "Let's now look into the medications data. We can isolate the medications data by searching the CONCEPT_PATH variable for the term `ACT Medications`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "758849f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>CONCEPT_PATH</th>\n",
       "      <th>NVAL_NUM</th>\n",
       "      <th>TVAL_CHAR</th>\n",
       "      <th>TIMESTAMP</th>\n",
       "      <th>VARIABLE_TYPE</th>\n",
       "      <th>DATE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2933374</th>\n",
       "      <td>3</td>\n",
       "      <td>\\ACT Medications\\I [Preparations]\\Ibuprofen\\Ibuprofen Oral Tablet\\Ibuprofen 100 Mg Oral Tablet</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2021-08-10 04:00:00</td>\n",
       "      <td>ACT Medications</td>\n",
       "      <td>2021-08-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2933375</th>\n",
       "      <td>4</td>\n",
       "      <td>\\ACT Medications\\D [Preparations]\\Diphenhydramine\\Diphenhydramine Oral Tablet\\Diphenhydramine Hydrochloride 25 Mg Oral Tablet</td>\n",
       "      <td>670.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1969-06-07 04:00:00</td>\n",
       "      <td>ACT Medications</td>\n",
       "      <td>1969-06-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2933376</th>\n",
       "      <td>4</td>\n",
       "      <td>\\ACT Medications\\C [Preparations]\\Clopidogrel\\Clopidogrel Oral Tablet\\Clopidogrel 75 Mg Oral Tablet</td>\n",
       "      <td>189.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2008-12-06 05:00:00</td>\n",
       "      <td>ACT Medications</td>\n",
       "      <td>2008-12-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2933377</th>\n",
       "      <td>4</td>\n",
       "      <td>\\ACT Medications\\S [Preparations]\\Simvastatin\\Simvastatin Oral Tablet\\Simvastatin 20 Mg Oral Tablet</td>\n",
       "      <td>189.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2008-12-06 05:00:00</td>\n",
       "      <td>ACT Medications</td>\n",
       "      <td>2008-12-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2933378</th>\n",
       "      <td>4</td>\n",
       "      <td>\\ACT Medications\\M [Preparations]\\Metoprolol\\Metoprolol Extended Release Oral Tablet\\24 Hr Metoprolol Succinate 100 Mg Extended Release Oral Tablet</td>\n",
       "      <td>189.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2008-12-06 05:00:00</td>\n",
       "      <td>ACT Medications</td>\n",
       "      <td>2008-12-06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         PATIENT_NUM  \\\n",
       "2933374            3   \n",
       "2933375            4   \n",
       "2933376            4   \n",
       "2933377            4   \n",
       "2933378            4   \n",
       "\n",
       "                                                                                                                                                CONCEPT_PATH  \\\n",
       "2933374                                                       \\ACT Medications\\I [Preparations]\\Ibuprofen\\Ibuprofen Oral Tablet\\Ibuprofen 100 Mg Oral Tablet   \n",
       "2933375                        \\ACT Medications\\D [Preparations]\\Diphenhydramine\\Diphenhydramine Oral Tablet\\Diphenhydramine Hydrochloride 25 Mg Oral Tablet   \n",
       "2933376                                                  \\ACT Medications\\C [Preparations]\\Clopidogrel\\Clopidogrel Oral Tablet\\Clopidogrel 75 Mg Oral Tablet   \n",
       "2933377                                                  \\ACT Medications\\S [Preparations]\\Simvastatin\\Simvastatin Oral Tablet\\Simvastatin 20 Mg Oral Tablet   \n",
       "2933378  \\ACT Medications\\M [Preparations]\\Metoprolol\\Metoprolol Extended Release Oral Tablet\\24 Hr Metoprolol Succinate 100 Mg Extended Release Oral Tablet   \n",
       "\n",
       "         NVAL_NUM TVAL_CHAR           TIMESTAMP    VARIABLE_TYPE        DATE  \n",
       "2933374       1.0       NaN 2021-08-10 04:00:00  ACT Medications  2021-08-10  \n",
       "2933375     670.0       NaN 1969-06-07 04:00:00  ACT Medications  1969-06-07  \n",
       "2933376     189.0       NaN 2008-12-06 05:00:00  ACT Medications  2008-12-06  \n",
       "2933377     189.0       NaN 2008-12-06 05:00:00  ACT Medications  2008-12-06  \n",
       "2933378     189.0       NaN 2008-12-06 05:00:00  ACT Medications  2008-12-06  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#creating a new dataframe containing only medication data:\n",
    "\n",
    "medications = data[data['CONCEPT_PATH'].str.contains(r'\\\\ACT Medications')]\n",
    "\n",
    "medications.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e75e6ad8",
   "metadata": {},
   "source": [
    "For our epilepsy case study, we are interested in knowing whether a patient has been administered Carbamazepine or Pregabalin. \n",
    "- Carbamazepine is used to manage and treat epilepsy, trigeminal neuralgia, and acute manic and mixed episodes in bipolar I disorder.\n",
    "- Pregabalin treats nerve and muscle pain, including fibromyalgia. It can also treat seizures.\n",
    "\n",
    "\n",
    "We will search our dataset for these medications and isolate which patients are associated to those observations.\n",
    "\n",
    "We then isolate our results to our cohort of interest. \n",
    "\n",
    "The resulting table contains information on whether or not a patient who was diagnosed with epilepsy was ever administered carbamazepine or pregabalin. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "a079b7de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>carbamazepine</th>\n",
       "      <th>pregabalin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>87</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>122</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>149</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>175</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PATIENT_NUM  carbamazepine  pregabalin\n",
       "0            9              0           0\n",
       "1           87              0           0\n",
       "2          122              0           0\n",
       "3          149              1           0\n",
       "4          175              0           0"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#filtering original data set to patients receiving Carbamazepine or/and Pregabalin\n",
    "\n",
    "carba_preg = (data[data['CONCEPT_PATH'].str.contains(r'Carbamazepine|Pregabalin')]\n",
    "              [['PATIENT_NUM', 'CONCEPT_PATH']])\n",
    "\n",
    "#creating a new column indicating if a patient has been prescribed Carbamazepine or Pregabalin (yes=1,no=0) for easier visualization and data manipulation \n",
    "carba_preg['carbamazepine'] = carba_preg['CONCEPT_PATH'].str.contains('Carbamazepine').astype(int)\n",
    "carba_preg['pregabalin'] = carba_preg['CONCEPT_PATH'].str.contains('Pregabalin').astype(int)\n",
    "\n",
    "carba_preg = carba_preg[['PATIENT_NUM', 'carbamazepine', 'pregabalin']].drop_duplicates()\n",
    "\n",
    "# merging the patient id list of epilepsy patients with the patient medication data frame, filing missing entries with \"0\"(=no/ no information included in dataset):\n",
    "\n",
    "patient_icd10_med = (\n",
    "    pd.merge(epilepsy_icd10_patient_ids, carba_preg, on='PATIENT_NUM', how='left')\n",
    "    .fillna(0)\n",
    "    .astype({'carbamazepine': 'int', 'pregabalin': 'int'})\n",
    ")\n",
    "\n",
    "\n",
    "patient_icd10_med.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "5d2a1e26",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "122 patient(s) with epilepsy were taking carbamazepine\n",
      "1 patient(s) with epilepsy were taking pregabalin\n"
     ]
    }
   ],
   "source": [
    "print(patient_icd10_med['carbamazepine'].value_counts()[1], \"patient(s) with epilepsy were taking carbamazepine\")\n",
    "print(patient_icd10_med['pregabalin'].value_counts()[1], \"patient(s) with epilepsy were taking pregabalin\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fcf39ac",
   "metadata": {},
   "source": [
    "### Invsetigating lab results over time for our cohort of interest\n",
    "\n",
    "We can isolate the lab related information by searching for 'ACT Lab Test Results' in the CONCEPT_PATH column.\n",
    "\n",
    "Below is a preview of the labs available in the dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "9db402d0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CONCEPT_PATH</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>\\ACT Lab Test Results\\Chemistry\\Electrolytes\\Bicarbonate\\Carbon dioxide in Serum or Plasma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>\\ACT Lab Test Results\\Chemistry\\Electrolytes\\Calcium\\Calcium in Blood</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\\ACT Lab Test Results\\Chemistry\\Electrolytes\\Calcium\\Calcium in Serum or Plasma</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>\\ACT Lab Test Results\\Chemistry\\Electrolytes\\Chloride\\Chloride in Blood</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>\\ACT Lab Test Results\\Chemistry\\Electrolytes\\Chloride\\Chloride in Serum or Plasma</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                 CONCEPT_PATH\n",
       "0  \\ACT Lab Test Results\\Chemistry\\Electrolytes\\Bicarbonate\\Carbon dioxide in Serum or Plasma\n",
       "1                       \\ACT Lab Test Results\\Chemistry\\Electrolytes\\Calcium\\Calcium in Blood\n",
       "2             \\ACT Lab Test Results\\Chemistry\\Electrolytes\\Calcium\\Calcium in Serum or Plasma\n",
       "3                     \\ACT Lab Test Results\\Chemistry\\Electrolytes\\Chloride\\Chloride in Blood\n",
       "4           \\ACT Lab Test Results\\Chemistry\\Electrolytes\\Chloride\\Chloride in Serum or Plasma"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating a new dataframe containing only the lab test results for all patients in the original dataset:\n",
    "lab_test_results = data[data['CONCEPT_PATH'].str.contains(r'\\\\ACT Lab Test Results')]\n",
    "\n",
    "# creating a list with unique types of lab tests included in the dataset:\n",
    "\n",
    "lab_list = (lab_test_results[['CONCEPT_PATH']]\n",
    "                   .drop_duplicates()\n",
    "                   .sort_values(by='CONCEPT_PATH')\n",
    "                   .reset_index(drop=True))\n",
    "\n",
    "lab_list.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0bf8737",
   "metadata": {},
   "source": [
    "In the following example, we will focus on AST (aspartate aminotransferase) and ALT (alanine transaminase) lab test results. \n",
    "We will also only focus on lab results from 2023. \n",
    "- AST is an enzyme found in the liver, muscles, and other organs. When cells that contain AST are damaged, they release AST into the blood\n",
    "- An ALT blood test is used to diagnose liver disorders. ALT is a liver enzyme that is released in the blood when liver cells are damaged.\n",
    "\n",
    "(you can replace \"ALT\" or \"AST\" with your choice of lab test, to add additional lab tests just add a \"|\" before the next lab test name)\n",
    "\n",
    "\n",
    "*Note that the number of lab result entries can vary widely per patient, and even multiple entries per day are possible. In the graphic representation and statistical measurement tables, each individual lab result is treated as an individual data point, without the use of weighted averages or other methodological statistical adjustments.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "967c6add",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>lab_test</th>\n",
       "      <th>value</th>\n",
       "      <th>TIMESTAMP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>145312</th>\n",
       "      <td>11</td>\n",
       "      <td>alt</td>\n",
       "      <td>40.8</td>\n",
       "      <td>2023-07-17 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145313</th>\n",
       "      <td>11</td>\n",
       "      <td>ast</td>\n",
       "      <td>35.2</td>\n",
       "      <td>2023-07-17 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146748</th>\n",
       "      <td>16</td>\n",
       "      <td>alt</td>\n",
       "      <td>49.6</td>\n",
       "      <td>2023-06-01 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146749</th>\n",
       "      <td>16</td>\n",
       "      <td>ast</td>\n",
       "      <td>33.3</td>\n",
       "      <td>2023-06-01 04:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149572</th>\n",
       "      <td>39</td>\n",
       "      <td>alt</td>\n",
       "      <td>52.0</td>\n",
       "      <td>2023-06-26 04:00:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        PATIENT_NUM lab_test  value           TIMESTAMP\n",
       "145312           11      alt   40.8 2023-07-17 04:00:00\n",
       "145313           11      ast   35.2 2023-07-17 04:00:00\n",
       "146748           16      alt   49.6 2023-06-01 04:00:00\n",
       "146749           16      ast   33.3 2023-06-01 04:00:00\n",
       "149572           39      alt   52.0 2023-06-26 04:00:00"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#filtering the original dataset to include only AST and ALT test results from the year 2023 and cleaning duplicates:\n",
    "alt_ast = lab_test_results[lab_test_results['CONCEPT_PATH'].str.contains('ALT|AST')]\n",
    "alt_ast = alt_ast[alt_ast['TIMESTAMP'] >= pd.to_datetime('2023-01-01')]\n",
    "alt_ast = alt_ast[alt_ast['TIMESTAMP'] <= pd.to_datetime('2023-12-31')]\n",
    "\n",
    "\n",
    "\n",
    "alt_ast['lab_test'] = np.where(\n",
    "    alt_ast['CONCEPT_PATH'].str.contains('AST'), 'ast',\n",
    "    np.where(alt_ast['CONCEPT_PATH'].str.contains('ALT'), 'alt', None)\n",
    ")\n",
    "\n",
    "\n",
    "alt_ast = (\n",
    "    alt_ast\n",
    "    .rename(columns={'NVAL_NUM': 'value'})\n",
    "    .loc[:, ['PATIENT_NUM', 'lab_test', 'value', 'TIMESTAMP']]\n",
    "    .sort_values(by=['PATIENT_NUM', 'TIMESTAMP'])\n",
    "    .drop_duplicates()\n",
    ")\n",
    "\n",
    "alt_ast.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abe62132",
   "metadata": {},
   "source": [
    "Let's now graph a distribution of ALT and AST values in 2023."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "3c0dfd99",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/nz/k_g_v4910fd_wthf2ngtf13c0000gq/T/ipykernel_38006/2280149245.py:5: FutureWarning: \n",
      "\n",
      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
      "\n",
      "  ax = sns.boxplot(x='lab_test', y='value', data=alt_ast, palette={'ast': 'skyblue', 'alt': 'red'})\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#creating a box plot graph of the results. \n",
    "\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "ax = sns.boxplot(x='lab_test', y='value', data=alt_ast, palette={'ast': 'skyblue', 'alt': 'red'})\n",
    "plt.title('Liver Transaminase Levels in 2023, n=' + str(len(alt_ast)))\n",
    "plt.xlabel('Lab Test')\n",
    "plt.ylabel('Value (u/l)')\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0fe3aec",
   "metadata": {},
   "source": [
    "Let's say we are interested in a specific patient with epilepsy. \n",
    "Below, we plot the AST and ALT lab trends over the year 2023 for patient id # 1f9e58a1-0b6f-7f53-1495-8b05b69ad4ed.\n",
    "\n",
    "You can modify this code for any patient of interest by changing the `patient_of_interest` variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "02c6d2cd",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         PATIENT_NUM lab_test  value           TIMESTAMP\n",
      "2931499        11641      alt   29.8 2023-01-23 05:00:00\n",
      "2931500        11641      ast   11.3 2023-01-23 05:00:00\n",
      "2931549        11641      alt   25.6 2023-02-23 05:00:00\n",
      "2931550        11641      ast   37.6 2023-02-23 05:00:00\n",
      "2931615        11641      alt   21.4 2023-03-27 04:00:00\n",
      "2931616        11641      ast   39.2 2023-03-27 04:00:00\n",
      "2931644        11641      alt   40.6 2023-04-27 04:00:00\n",
      "2931645        11641      ast   25.7 2023-04-27 04:00:00\n",
      "2931686        11641      alt   34.1 2023-05-28 04:00:00\n",
      "2931687        11641      ast   15.7 2023-05-28 04:00:00\n",
      "2932099        11641      alt   41.3 2023-10-22 04:00:00\n",
      "2932100        11641      ast   33.9 2023-10-22 04:00:00\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "patient_of_interest = 11641\n",
    "\n",
    "#filtering for the specific patient's AST and ALT levels over time using the 2023 alt/ast dataframe we previously created:\n",
    "single_patient_data = alt_ast[alt_ast['PATIENT_NUM'] == patient_of_interest]\n",
    "print(single_patient_data)\n",
    "\n",
    "# Pivot the data so that each lab test becomes a separate column\n",
    "patient_data_pivot = single_patient_data.pivot(index='TIMESTAMP', columns='lab_test', values='value')\n",
    "\n",
    "#creaing a plot showing the results over time:\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(patient_data_pivot.index, patient_data_pivot['alt'], marker='o', linestyle='-', color='r', label='alt')\n",
    "plt.plot(patient_data_pivot.index, patient_data_pivot['ast'], marker='o', linestyle='-', color='b', label='ast')\n",
    "plt.title('Liver transaminase levels over time for patient ' +  str(patient_of_interest) + ' in 2023')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Value (u/l)')\n",
    "plt.xticks(rotation=45)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dc7ea78",
   "metadata": {},
   "source": [
    "## Investigating procedural data over time. Case study: Appendectomy.\n",
    "\n",
    "Finally, let's take a look at the procedures data. \n",
    "\n",
    "The procedures are defined by CPT codes. \n",
    "We will explore the procedures in this data set and look at specific procedures of interest in a given time frame.\n",
    "\n",
    "We can isolate the procedure related information by searching for 'ACT Procedures' in the CONCEPT_PATH column.\n",
    "\n",
    "Below is an example of the procedures documented in this dataset.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "741f166c",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CONCEPT_PATH</th>\n",
       "      <th>procedure</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3682095</th>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>Radiologic examination, chest</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3682096</th>\n",
       "      <td>\\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest</td>\n",
       "      <td>Radiologic examination</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3682097</th>\n",
       "      <td>\\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Digestive System\\Surgical Procedures on the Colon and Rectum</td>\n",
       "      <td>Endoscopy Procedures on the Rectum</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3682104</th>\n",
       "      <td>\\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Integumentary System\\Surgical Procedures on the Breast\\Excision Procedures on the Breast</td>\n",
       "      <td>Biopsy of breast</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3682106</th>\n",
       "      <td>\\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Female Genital System\\Surgical Procedures on the Corpus Uteri\\Introduction Procedures on the Corpus Uteri</td>\n",
       "      <td>Insertion of intrauterine device (IUD)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                             CONCEPT_PATH  \\\n",
       "3682095   \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3682096   \\ACT Procedures CPT\\Radiology Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures\\Diagnostic Radiology (Diagnostic Imaging) Procedures of the Chest   \n",
       "3682097                                               \\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Digestive System\\Surgical Procedures on the Colon and Rectum   \n",
       "3682104                   \\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Integumentary System\\Surgical Procedures on the Breast\\Excision Procedures on the Breast   \n",
       "3682106  \\ACT Procedures CPT\\Surgery\\Surgical Procedures on the Female Genital System\\Surgical Procedures on the Corpus Uteri\\Introduction Procedures on the Corpus Uteri   \n",
       "\n",
       "                                      procedure  \n",
       "3682095           Radiologic examination, chest  \n",
       "3682096                  Radiologic examination  \n",
       "3682097      Endoscopy Procedures on the Rectum  \n",
       "3682104                        Biopsy of breast  \n",
       "3682106  Insertion of intrauterine device (IUD)  "
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creating a list with unique types of procedures included in the dataset :\n",
    "\n",
    "procedures = (\n",
    "    data[data['CONCEPT_PATH'].str.contains(r'\\\\ACT Procedures')]\n",
    "    .loc[:, ['CONCEPT_PATH', 'TVAL_CHAR']]\n",
    "    .drop_duplicates()\n",
    "    .rename(columns={'TVAL_CHAR': 'procedure'})\n",
    ")\n",
    "\n",
    "procedures.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96a52f1c",
   "metadata": {},
   "source": [
    "Below, we filter the procedural data to identify a cohort of patients who have undergone an appendectomy since the beginning of 2020."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "1cc76554",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PATIENT_NUM</th>\n",
       "      <th>procedure</th>\n",
       "      <th>TIMESTAMP</th>\n",
       "      <th>VARIABLE_TYPE</th>\n",
       "      <th>DATE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3683372</th>\n",
       "      <td>251</td>\n",
       "      <td>Appendectomy</td>\n",
       "      <td>2020-11-18 05:00:00</td>\n",
       "      <td>ACT Procedures CPT</td>\n",
       "      <td>2020-11-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3687591</th>\n",
       "      <td>1210</td>\n",
       "      <td>Appendectomy</td>\n",
       "      <td>2021-11-18 05:00:00</td>\n",
       "      <td>ACT Procedures CPT</td>\n",
       "      <td>2021-11-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3690114</th>\n",
       "      <td>1743</td>\n",
       "      <td>Appendectomy</td>\n",
       "      <td>2022-11-28 05:00:00</td>\n",
       "      <td>ACT Procedures CPT</td>\n",
       "      <td>2022-11-28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3690148</th>\n",
       "      <td>1759</td>\n",
       "      <td>Appendectomy</td>\n",
       "      <td>2021-08-31 04:00:00</td>\n",
       "      <td>ACT Procedures CPT</td>\n",
       "      <td>2021-08-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3691958</th>\n",
       "      <td>2192</td>\n",
       "      <td>Appendectomy</td>\n",
       "      <td>2020-10-17 04:00:00</td>\n",
       "      <td>ACT Procedures CPT</td>\n",
       "      <td>2020-10-17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         PATIENT_NUM     procedure           TIMESTAMP       VARIABLE_TYPE  \\\n",
       "3683372          251  Appendectomy 2020-11-18 05:00:00  ACT Procedures CPT   \n",
       "3687591         1210  Appendectomy 2021-11-18 05:00:00  ACT Procedures CPT   \n",
       "3690114         1743  Appendectomy 2022-11-28 05:00:00  ACT Procedures CPT   \n",
       "3690148         1759  Appendectomy 2021-08-31 04:00:00  ACT Procedures CPT   \n",
       "3691958         2192  Appendectomy 2020-10-17 04:00:00  ACT Procedures CPT   \n",
       "\n",
       "               DATE  \n",
       "3683372  2020-11-18  \n",
       "3687591  2021-11-18  \n",
       "3690114  2022-11-28  \n",
       "3690148  2021-08-31  \n",
       "3691958  2020-10-17  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filtering to create a new dataframe including patients with a history of an appendectomy procedure\n",
    "\n",
    "appendectomy = (\n",
    "    data[data['TVAL_CHAR'].str.contains('Appendectomy',na=False)]\n",
    "    .rename(columns={'TVAL_CHAR': 'procedure'})\n",
    "    .drop(columns=['CONCEPT_PATH', 'NVAL_NUM'])\n",
    "    .drop_duplicates()\n",
    ")\n",
    "\n",
    "# filtering to only include patients who had an appendectomy since the beginning of the year 2020\n",
    "\n",
    "appendectomy_2020 = appendectomy[appendectomy['TIMESTAMP'] >= pd.to_datetime('2020-01-01')]\n",
    "\n",
    "appendectomy_2020.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64c711cb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}