FinalProject_group077.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COGS 108 - Final Project "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Overview"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The question our project is addressing is whether the population of a state affects the amount of medicare one receives for a particular illness. This topic is of great importance because medicare provides health insurance to seniors, the disabled and those who are unable to afford it. Our project aims to find a relation between population and Medicare to tell if the state you live in has any disadvantage when it comes to Medicare, and whether it differs for different illnesses."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Names\n",
    "\n",
    "- Samvit Sutodiya\n",
    "- Arya Nair\n",
    "- Linwei Lu\n",
    "- Jenny Zhang"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Research Question"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The question our project seeks an answer to is whether the population of a state affect the amount of medicare one receives for a particular illness. We analyze state populations and average medicare coverage for specific diseases in each state. We compare the amount covered to the amount paid and see if population has any effect on that ratio."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Background and Prior Work"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Medicare is a national health insurance program that covers health care costs for adults over the age of 65 and younger people with specific disabilities. Medicare has four parts that help cover certain services such as Hospital Insurance, Medical Insurance, prescription drug coverage and a Medicare Advantage which is an all in one plan. The main benefit of Medicare is that it provides health coverage regardless of the person’s income, thus benefiting people with low income by helping those who are unable to afford medical insurance (“Acute Inpatient PPS”, 2019).\n",
    "\n",
    "But unfortunately this isn’t always the case. Studies such as ‘Medicare Utilization in the United States: PSRO and Regional Impacts’ by DL Rothberg, J Pinto and PM Gertman show how geographical factors affect medicare in the United States. Factors such as demographic and socio-economics play a role in determining the amount of medicare one receives. In this project we explore how population plays a role in determining the amount of medicare one receives. \n",
    "\n",
    "References (include links):\n",
    "\n",
    "- “Acute Inpatient PPS”, 2019 (https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/index.html?redirect=/AcuteInpatientPPS/)\n",
    "- ‘Medicare Utilization in the United States: PSRO and Regional Impacts’ by DL Rothberg, J Pinto and PM Gertman (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1072167/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hypothesis\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The question our project asks is whether the amount of medicare one receives for a specific illness, differs from state to state based on the state’s population. The amount of medicare is calculated as the ratio of amount covered by medicare to the amount paid by patients.\n",
    "\n",
    "The hypothesis we have is that the greater the population of the state, the more medicare one receives on average. Due to the higher populations, more people would be paying more premiums and drive up the coverage provided by health insurance companies. This might result in a higher coverage for people in these states. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Datasets- Description"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this project, we utilize three Datasets: one relating to the medical insurance and two involving the state population details. Given on these three data sets, we would able to found out \"whether the amount of medicare one receives for a specific illness, differs from state to state based on the state’s population\"(project proposal). \n",
    "\n",
    "First dataset we will involve is the \"Inpatient Prospective Payment System (IPPS) Provider Summary for the Top 100 Diagnosis-Related Groups (DRG) - FY2011\". The Inpatient Prospective Payment System(MIPPS) is a payment system for inpatient care and stays covered by medical insurance (Medicare). The data table includes Total Discharges, Average Covered Discharges, Average Total Payments, and Average Medicare Payments which help extract the amount of Medicare provided for a specific illness. what's more, the data set provides over 3,000 u.s hospitals spefical charges information, and top 100 most frequently billed discharges involving with the MIPPS payments. In the columns of the dataset, it involves DRG to represents over \"7 million discharges or 60 percent of totalMedicare IPPS discharges\". It classifies the system grouping similar clinical conditions and procedures and the provider city. Hence, the said data tables help us to answer the question of how much the patients pay for the different illness in each state.\n",
    "\n",
    "- DataSet Name: \"Inpatient Prospective Payment System (IPPS) Provider Summary for the Top 100 Diagnosis-Related Groups (DRG) - FY2011\".\n",
    "- Link to the dataset: https://data.cms.gov/Medicare-Inpatient/Inpatient-Prospective-Payment-System-IPPS-Provider/97k6-zzx3\n",
    "- Data Provided by CMs, DataSet Owner OEDA\n",
    "- Number of observations: 163065\n",
    "\n",
    "Second dataset we included is the \"National, State, and Puerto Rico Commonwealth Totals Datasets: Population, population change, and estimated components of population change: April 1, 2010 to July 1, 2018\". It is provided by United States Census Bureau. This dataset contains data from the 2010 US Census, as well as estimated yearly population, population change, and estimated components of population change of each state from 2010-2018.In the columns of table, it includes Name of the Region, Popluation Estimate ,National Population Change, Birth Rate, Death Rate, which provides the conprehensive information of the population changes in demestic and national scales. \n",
    "\n",
    "- DataSet Name:\"National, State, and Puerto Rico Commonwealth Totals Datasets: Population, population change, and estimated components of population change: April 1, 2010 to July 1, 2018\". \n",
    "- Link to the dataset: https://www.census.gov/data/datasets/time-series/demo/popest/2010s-state-total.html \n",
    "- Data Provided by United States Census Bureau\n",
    "- Number of observations: 57\n",
    "\n",
    "\n",
    "The third one is \"CSV of States and State Abbreviations\", provided by fronz.net weblog. This dataset contains all the US states and their abbreviation.Unlike the said two datasets, this dataset table is straightforward: there are only two columnes: State name and Abbreviation. This dataset  increase the efficency to recognize, categorize and label the geographic information.\n",
    "\n",
    "- DataSet Name: \"CSV of States and State Abbreviations\".\n",
    "- Link to the dataset: http://www.fonz.net/blog/archives/2008/04/06/csv-of-states-and-state-abbreviations/\n",
    "- Number of observations: 51\n",
    "\n",
    "In conclusion, we plan to combine these three databases by adding the populations for each provider state to analyze whether the population of a state affects the amount of Medicare one receives. Since the first dataset uses state abbreviations while the census uses their names, we will be using a third dataset to match abbreviations to state names. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import patsy\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.api as sm\n",
    "plt.rcParams['figure.figsize'] = (17, 7) \n",
    "import seaborn as sns\n",
    "sns.set(style='white', font_scale=2) \n",
    "from scipy.stats import uniform, norm, bernoulli, poisson, ttest_ind"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we will read in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = 'Inpatient_Prospective_Payment_System__IPPS__Provider_Summary_' \\\n",
    "            'for_the_Top_100_Diagnosis-Related_Groups__DRG__-_FY2011.csv'\n",
    "df = pd.read_csv(dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we will drop the columns that we won't be using in our analysis, which are all the provider-related fields (aside from the state) and the hospital region. Some of the column names have extra spaces around them, so we will rename those columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(columns = ['Provider Id', 'Provider Name', 'Provider Street Address', \n",
    "                   'Provider City', 'Provider Zip Code',\n",
    "                   'Hospital Referral Region Description'],\n",
    "                 inplace = True)\n",
    "columns = ['DRG', 'State', 'Discharges', 'Avg Covered Charges',\n",
    "           'Avg Total Payments', 'Avg Medicare Payments']\n",
    "df.columns = columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to combine all the data for each state. We will be analyzing different illnesses/conditions separately, so each state will have one row for each illness containing the Average Covered Charges (the provider's average charge for services) and Average Medicare Payments (the amount covered by Medicare, not including co-payment and deductibles). We will keep the Total Payments column as well, in case we want to analyze it separately later. Using the number of total discharges from each hospital, we will calculate each of these average amounts for each illness within each state.\n",
    "\n",
    "Using these average values, we will create two new dataframes to perform analysis on. The first dataframe will contain the average amounts for each state and condition. The second dataframe will only contain the average values for each state, for the purpose of simplifying our overall analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# construct the rows of the first dataframe, separated by condition and state\n",
    "rows_condition = list()\n",
    "for name, group in df.groupby(['State', 'DRG']):\n",
    "    total_discharges = group['Discharges'].sum()\n",
    "    avg_charges = (group['Avg Covered Charges'] * group['Discharges']).sum()\n",
    "    / total_discharges\n",
    "    avg_total = (group['Avg Total Payments'] * group['Discharges']).sum()\n",
    "    / total_discharges\n",
    "    avg_medicare = (group['Avg Medicare Payments'] * group['Discharges']).sum()\n",
    "    / total_discharges\n",
    "    rows_condition.append([name[1], name[0], total_discharges, avg_charges,\n",
    "                           avg_total, avg_medicare])\n",
    "\n",
    "# construct the rows of the second dataframe, separated by state\n",
    "rows_state = list()\n",
    "for name, group in df.groupby(['State']):\n",
    "    total_discharges = group['Discharges'].sum()\n",
    "    avg_charges = (group['Avg Covered Charges'] * group['Discharges']).sum() \n",
    "    / total_discharges\n",
    "    avg_total = (group['Avg Total Payments'] * group['Discharges']).sum()\n",
    "    / total_discharges\n",
    "    avg_medicare = (group['Avg Medicare Payments'] * group['Discharges']).sum()\n",
    "    / total_discharges\n",
    "    rows_state.append([name, total_discharges, avg_charges, avg_total,\n",
    "                       avg_medicare])\n",
    "    \n",
    "df_condition = pd.DataFrame(rows_condition, columns = columns)\n",
    "df_state = pd.DataFrame(rows_state, columns = ['State', 'Discharges',\n",
    "                                               'Avg Covered Charges',\n",
    "                                               'Avg Total Payments',\n",
    "                                               'Avg Medicare Payments'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our df_condition dataframe currently contains data for the top 100 Diagnosis-Related Groups (DRG). To simplify our analysis, we will narrow down our data to the top 10 DRG, based on number of total discharges."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "rows_DRG = list()\n",
    "for name, group in df.groupby(['DRG']):\n",
    "    rows_DRG.append([name, group['Discharges'].sum()])\n",
    "df_DRG = pd.DataFrame(rows_DRG, columns = ['DRG', 'Discharges'])\n",
    "top = df_DRG.nlargest(10, 'Discharges')['DRG']\n",
    "df_condition = df_condition[df_condition['DRG'].isin(top)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we will read in the state population dataset. The population dataset uses the full names of states, but our Medicare dataset uses their abbreviations. To match populations with states, we will use a third dataset containing the abbreviations of states to add the abbreviations to the population dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_population = pd.read_csv('nst-est2018-alldata.csv')\n",
    "df_states = pd.read_csv('states.csv')\n",
    "\n",
    "# drop all columns except the two necessary columns\n",
    "df_population = df_population[['NAME', 'POPESTIMATE2018']]\n",
    "\n",
    "# rename the columns to merge with df_states\n",
    "df_population.columns = ['State', 'Population']\n",
    "df_population = df_population.merge(df_states, on='State')\n",
    "\n",
    "# rename the columns again to merge with Medicare dataframes\n",
    "df_population.drop(columns = ['State'], inplace = True)\n",
    "df_population.columns = ['Population', 'State']\n",
    "\n",
    "df_condition = df_condition.merge(df_population, on='State')\n",
    "df_state = df_state.merge(df_population, on='State')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we will add a column for the ratio of the amount covered by medicare to the total charges from the provider."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_state['Avg % Covered'] = df_state['Avg Medicare Payments'] \\\n",
    "                            / df_state['Avg Covered Charges']\n",
    "df_condition['Avg % Covered'] = df_condition['Avg Medicare Payments'] \\\n",
    "                            / df_condition['Avg Covered Charges']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before performing targeted analysis, we must perform some exploratory data analysis, to understand the dataset better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>State</th>\n",
       "      <th>Discharges</th>\n",
       "      <th>Avg Covered Charges</th>\n",
       "      <th>Avg Total Payments</th>\n",
       "      <th>Avg Medicare Payments</th>\n",
       "      <th>Population</th>\n",
       "      <th>Avg % Covered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AK</td>\n",
       "      <td>6142</td>\n",
       "      <td>2.638110e+08</td>\n",
       "      <td>9.014987e+07</td>\n",
       "      <td>7.901020e+07</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.299496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AL</td>\n",
       "      <td>142704</td>\n",
       "      <td>4.579079e+09</td>\n",
       "      <td>1.098627e+09</td>\n",
       "      <td>9.285877e+08</td>\n",
       "      <td>4887871</td>\n",
       "      <td>0.202789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AR</td>\n",
       "      <td>86769</td>\n",
       "      <td>2.191450e+09</td>\n",
       "      <td>6.824737e+08</td>\n",
       "      <td>5.845230e+08</td>\n",
       "      <td>3013825</td>\n",
       "      <td>0.266729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>104604</td>\n",
       "      <td>4.127849e+09</td>\n",
       "      <td>1.023640e+09</td>\n",
       "      <td>8.820153e+08</td>\n",
       "      <td>7171646</td>\n",
       "      <td>0.213674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA</td>\n",
       "      <td>474979</td>\n",
       "      <td>3.160585e+10</td>\n",
       "      <td>5.862212e+09</td>\n",
       "      <td>5.314672e+09</td>\n",
       "      <td>39557045</td>\n",
       "      <td>0.168155</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  State  Discharges  Avg Covered Charges  Avg Total Payments  \\\n",
       "0    AK        6142         2.638110e+08        9.014987e+07   \n",
       "1    AL      142704         4.579079e+09        1.098627e+09   \n",
       "2    AR       86769         2.191450e+09        6.824737e+08   \n",
       "3    AZ      104604         4.127849e+09        1.023640e+09   \n",
       "4    CA      474979         3.160585e+10        5.862212e+09   \n",
       "\n",
       "   Avg Medicare Payments  Population  Avg % Covered  \n",
       "0           7.901020e+07      737438       0.299496  \n",
       "1           9.285877e+08     4887871       0.202789  \n",
       "2           5.845230e+08     3013825       0.266729  \n",
       "3           8.820153e+08     7171646       0.213674  \n",
       "4           5.314672e+09    39557045       0.168155  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>DRG</th>\n",
       "      <th>State</th>\n",
       "      <th>Discharges</th>\n",
       "      <th>Avg Covered Charges</th>\n",
       "      <th>Avg Total Payments</th>\n",
       "      <th>Avg Medicare Payments</th>\n",
       "      <th>Population</th>\n",
       "      <th>Avg % Covered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>190 - CHRONIC OBSTRUCTIVE PULMONARY DISEASE W MCC</td>\n",
       "      <td>AK</td>\n",
       "      <td>163</td>\n",
       "      <td>5615063.50</td>\n",
       "      <td>2099027.40</td>\n",
       "      <td>1889600.38</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.336523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>194 - SIMPLE PNEUMONIA &amp; PLEURISY W CC</td>\n",
       "      <td>AK</td>\n",
       "      <td>320</td>\n",
       "      <td>7720865.07</td>\n",
       "      <td>3453228.26</td>\n",
       "      <td>3136002.95</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.406172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>291 - HEART FAILURE &amp; SHOCK W MCC</td>\n",
       "      <td>AK</td>\n",
       "      <td>111</td>\n",
       "      <td>5507783.25</td>\n",
       "      <td>1770687.53</td>\n",
       "      <td>1678110.53</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.304680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>292 - HEART FAILURE &amp; SHOCK W CC</td>\n",
       "      <td>AK</td>\n",
       "      <td>217</td>\n",
       "      <td>6956941.48</td>\n",
       "      <td>2412829.24</td>\n",
       "      <td>2065336.15</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.296874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>392 - ESOPHAGITIS, GASTROENT &amp; MISC DIGEST DIS...</td>\n",
       "      <td>AK</td>\n",
       "      <td>210</td>\n",
       "      <td>4068630.15</td>\n",
       "      <td>1626559.12</td>\n",
       "      <td>1354903.06</td>\n",
       "      <td>737438</td>\n",
       "      <td>0.333012</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 DRG State  Discharges  \\\n",
       "0  190 - CHRONIC OBSTRUCTIVE PULMONARY DISEASE W MCC    AK         163   \n",
       "1             194 - SIMPLE PNEUMONIA & PLEURISY W CC    AK         320   \n",
       "2                  291 - HEART FAILURE & SHOCK W MCC    AK         111   \n",
       "3                   292 - HEART FAILURE & SHOCK W CC    AK         217   \n",
       "4  392 - ESOPHAGITIS, GASTROENT & MISC DIGEST DIS...    AK         210   \n",
       "\n",
       "   Avg Covered Charges  Avg Total Payments  Avg Medicare Payments  Population  \\\n",
       "0           5615063.50          2099027.40             1889600.38      737438   \n",
       "1           7720865.07          3453228.26             3136002.95      737438   \n",
       "2           5507783.25          1770687.53             1678110.53      737438   \n",
       "3           6956941.48          2412829.24             2065336.15      737438   \n",
       "4           4068630.15          1626559.12             1354903.06      737438   \n",
       "\n",
       "   Avg % Covered  \n",
       "0       0.336523  \n",
       "1       0.406172  \n",
       "2       0.304680  \n",
       "3       0.296874  \n",
       "4       0.333012  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 51 entries, 0 to 50\n",
      "Data columns (total 7 columns):\n",
      "State                    51 non-null object\n",
      "Discharges               51 non-null int64\n",
      "Avg Covered Charges      51 non-null float64\n",
      "Avg Total Payments       51 non-null float64\n",
      "Avg Medicare Payments    51 non-null float64\n",
      "Population               51 non-null int64\n",
      "Avg % Covered            51 non-null float64\n",
      "dtypes: float64(4), int64(2), object(1)\n",
      "memory usage: 3.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df_state.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 510 entries, 0 to 509\n",
      "Data columns (total 8 columns):\n",
      "DRG                      510 non-null object\n",
      "State                    510 non-null object\n",
      "Discharges               510 non-null int64\n",
      "Avg Covered Charges      510 non-null float64\n",
      "Avg Total Payments       510 non-null float64\n",
      "Avg Medicare Payments    510 non-null float64\n",
      "Population               510 non-null int64\n",
      "Avg % Covered            510 non-null float64\n",
      "dtypes: float64(4), int64(2), object(2)\n",
      "memory usage: 35.9+ KB\n"
     ]
    }
   ],
   "source": [
    "df_condition.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our data does not have any null values, which makes it easier to work with."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Discharges</th>\n",
       "      <th>Avg Covered Charges</th>\n",
       "      <th>Avg Total Payments</th>\n",
       "      <th>Avg Medicare Payments</th>\n",
       "      <th>Population</th>\n",
       "      <th>Avg % Covered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>5.100000e+01</td>\n",
       "      <td>5.100000e+01</td>\n",
       "      <td>5.100000e+01</td>\n",
       "      <td>5.100000e+01</td>\n",
       "      <td>51.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>136770.941176</td>\n",
       "      <td>4.823096e+09</td>\n",
       "      <td>1.307637e+09</td>\n",
       "      <td>1.136361e+09</td>\n",
       "      <td>6.415048e+06</td>\n",
       "      <td>0.289540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>136373.001344</td>\n",
       "      <td>6.326048e+09</td>\n",
       "      <td>1.372992e+09</td>\n",
       "      <td>1.211974e+09</td>\n",
       "      <td>7.343308e+06</td>\n",
       "      <td>0.107716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>6142.000000</td>\n",
       "      <td>2.027677e+08</td>\n",
       "      <td>7.899953e+07</td>\n",
       "      <td>6.667565e+07</td>\n",
       "      <td>5.777370e+05</td>\n",
       "      <td>0.143390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>27672.000000</td>\n",
       "      <td>7.667604e+08</td>\n",
       "      <td>2.710481e+08</td>\n",
       "      <td>2.346241e+08</td>\n",
       "      <td>1.780020e+06</td>\n",
       "      <td>0.234108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>96258.000000</td>\n",
       "      <td>2.763332e+09</td>\n",
       "      <td>9.172638e+08</td>\n",
       "      <td>7.799061e+08</td>\n",
       "      <td>4.468402e+06</td>\n",
       "      <td>0.283327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>190153.500000</td>\n",
       "      <td>5.475181e+09</td>\n",
       "      <td>1.707302e+09</td>\n",
       "      <td>1.445503e+09</td>\n",
       "      <td>7.353618e+06</td>\n",
       "      <td>0.324085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>536859.000000</td>\n",
       "      <td>3.160585e+10</td>\n",
       "      <td>5.862212e+09</td>\n",
       "      <td>5.314672e+09</td>\n",
       "      <td>3.955704e+07</td>\n",
       "      <td>0.853637</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Discharges  Avg Covered Charges  Avg Total Payments  \\\n",
       "count      51.000000         5.100000e+01        5.100000e+01   \n",
       "mean   136770.941176         4.823096e+09        1.307637e+09   \n",
       "std    136373.001344         6.326048e+09        1.372992e+09   \n",
       "min      6142.000000         2.027677e+08        7.899953e+07   \n",
       "25%     27672.000000         7.667604e+08        2.710481e+08   \n",
       "50%     96258.000000         2.763332e+09        9.172638e+08   \n",
       "75%    190153.500000         5.475181e+09        1.707302e+09   \n",
       "max    536859.000000         3.160585e+10        5.862212e+09   \n",
       "\n",
       "       Avg Medicare Payments    Population  Avg % Covered  \n",
       "count           5.100000e+01  5.100000e+01      51.000000  \n",
       "mean            1.136361e+09  6.415048e+06       0.289540  \n",
       "std             1.211974e+09  7.343308e+06       0.107716  \n",
       "min             6.667565e+07  5.777370e+05       0.143390  \n",
       "25%             2.346241e+08  1.780020e+06       0.234108  \n",
       "50%             7.799061e+08  4.468402e+06       0.283327  \n",
       "75%             1.445503e+09  7.353618e+06       0.324085  \n",
       "max             5.314672e+09  3.955704e+07       0.853637  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Discharges</th>\n",
       "      <th>Avg Covered Charges</th>\n",
       "      <th>Avg Total Payments</th>\n",
       "      <th>Avg Medicare Payments</th>\n",
       "      <th>Population</th>\n",
       "      <th>Avg % Covered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>510.000000</td>\n",
       "      <td>5.100000e+02</td>\n",
       "      <td>5.100000e+02</td>\n",
       "      <td>5.100000e+02</td>\n",
       "      <td>5.100000e+02</td>\n",
       "      <td>510.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4426.737255</td>\n",
       "      <td>1.471053e+08</td>\n",
       "      <td>4.071712e+07</td>\n",
       "      <td>3.523568e+07</td>\n",
       "      <td>6.415048e+06</td>\n",
       "      <td>0.295430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4902.784534</td>\n",
       "      <td>2.741529e+08</td>\n",
       "      <td>6.293694e+07</td>\n",
       "      <td>5.558612e+07</td>\n",
       "      <td>7.278097e+06</td>\n",
       "      <td>0.112333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>80.000000</td>\n",
       "      <td>1.890407e+06</td>\n",
       "      <td>8.395538e+05</td>\n",
       "      <td>6.454178e+05</td>\n",
       "      <td>5.777370e+05</td>\n",
       "      <td>0.115372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>853.500000</td>\n",
       "      <td>1.857804e+07</td>\n",
       "      <td>6.711661e+06</td>\n",
       "      <td>5.678361e+06</td>\n",
       "      <td>1.754208e+06</td>\n",
       "      <td>0.230998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2803.000000</td>\n",
       "      <td>5.928080e+07</td>\n",
       "      <td>1.920324e+07</td>\n",
       "      <td>1.640383e+07</td>\n",
       "      <td>4.468402e+06</td>\n",
       "      <td>0.280194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6122.750000</td>\n",
       "      <td>1.515708e+08</td>\n",
       "      <td>4.736295e+07</td>\n",
       "      <td>4.042992e+07</td>\n",
       "      <td>7.535591e+06</td>\n",
       "      <td>0.333011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>34284.000000</td>\n",
       "      <td>3.089759e+09</td>\n",
       "      <td>5.741585e+08</td>\n",
       "      <td>5.403977e+08</td>\n",
       "      <td>3.955704e+07</td>\n",
       "      <td>0.886906</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Discharges  Avg Covered Charges  Avg Total Payments  \\\n",
       "count    510.000000         5.100000e+02        5.100000e+02   \n",
       "mean    4426.737255         1.471053e+08        4.071712e+07   \n",
       "std     4902.784534         2.741529e+08        6.293694e+07   \n",
       "min       80.000000         1.890407e+06        8.395538e+05   \n",
       "25%      853.500000         1.857804e+07        6.711661e+06   \n",
       "50%     2803.000000         5.928080e+07        1.920324e+07   \n",
       "75%     6122.750000         1.515708e+08        4.736295e+07   \n",
       "max    34284.000000         3.089759e+09        5.741585e+08   \n",
       "\n",
       "       Avg Medicare Payments    Population  Avg % Covered  \n",
       "count           5.100000e+02  5.100000e+02     510.000000  \n",
       "mean            3.523568e+07  6.415048e+06       0.295430  \n",
       "std             5.558612e+07  7.278097e+06       0.112333  \n",
       "min             6.454178e+05  5.777370e+05       0.115372  \n",
       "25%             5.678361e+06  1.754208e+06       0.230998  \n",
       "50%             1.640383e+07  4.468402e+06       0.280194  \n",
       "75%             4.042992e+07  7.535591e+06       0.333011  \n",
       "max             5.403977e+08  3.955704e+07       0.886906  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "sns.heatmap(df_state.corr(), annot = True, cmap=\"YlGnBu\", ax = ax)\n",
    "ax.title.set_text('df_state Corellation Heatmap')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Population is correlated to Total Discharges (0.91)\n",
    "* Average Medicare Payments is highly correlated to Average Total Payments (0.99)\n",
    "* There is a slight correlation between Population and Average Covered Charges (0.52)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "sns.heatmap(df_condition.corr(), annot = True, cmap=\"YlGnBu\", ax = ax)\n",
    "ax.title.set_text('df_condition Corellation Heatmap')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* There is a slight corellation between Population and Average Covered Charges (0.35) in this dataset as well"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis & Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51, 7)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(510, 8)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 7)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "S_rows = df_state.isnull().any(axis=1)\n",
    "df_state[S_rows].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 8)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "C_rows = df_condition.isnull().any(axis=1)\n",
    "df_condition[C_rows].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([], dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.columns[df_state.isnull().any(axis=0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([], dtype='object')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.columns[df_condition.isnull().any(axis=0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "State                    0\n",
       "Discharges               0\n",
       "Avg Covered Charges      0\n",
       "Avg Total Payments       0\n",
       "Avg Medicare Payments    0\n",
       "Population               0\n",
       "Avg % Covered            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "State                    0\n",
       "Discharges               0\n",
       "Avg Covered Charges      0\n",
       "Avg Total Payments       0\n",
       "Avg Medicare Payments    0\n",
       "Population               0\n",
       "Avg % Covered            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "MEDIAA, MEAA, RANGE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Discharges               9.625800e+04\n",
       "Avg Covered Charges      2.763332e+09\n",
       "Avg Total Payments       9.172638e+08\n",
       "Avg Medicare Payments    7.799061e+08\n",
       "Population               4.468402e+06\n",
       "Avg % Covered            2.833274e-01\n",
       "dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Discharges               2.803000e+03\n",
       "Avg Covered Charges      5.928080e+07\n",
       "Avg Total Payments       1.920324e+07\n",
       "Avg Medicare Payments    1.640383e+07\n",
       "Population               4.468402e+06\n",
       "Avg % Covered            2.801941e-01\n",
       "dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Discharges               1.367709e+05\n",
       "Avg Covered Charges      4.823096e+09\n",
       "Avg Total Payments       1.307637e+09\n",
       "Avg Medicare Payments    1.136361e+09\n",
       "Population               6.415048e+06\n",
       "Avg % Covered            2.895403e-01\n",
       "dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_state.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Discharges               4.426737e+03\n",
       "Avg Covered Charges      1.471053e+08\n",
       "Avg Total Payments       4.071712e+07\n",
       "Avg Medicare Payments    3.523568e+07\n",
       "Population               6.415048e+06\n",
       "Avg % Covered            2.954297e-01\n",
       "dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_condition.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8536372030272014 - 0.14338960302819403 = 0.7102475999990073\n",
      "536859 - 6142 = 530717\n"
     ]
    }
   ],
   "source": [
    "s_max=df_state['Avg % Covered'].max()\n",
    "s_min=df_state['Avg % Covered'].min()\n",
    "s_range= s_max-s_min\n",
    "print(s_max,'-',s_min,'=',s_range)\n",
    "\n",
    "s_max1=df_state['Discharges'].max()\n",
    "s_min1=df_state['Discharges'].min()\n",
    "s_range1= s_max1-s_min1\n",
    "print(s_max1,'-',s_min1,'=',s_range1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "IQR(interquartile Range)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(6142.0, 536859.0)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "low, up = np.percentile(df_state['Discharges'], [0,100])\n",
    "low, up"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "530717.0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iqr = up - low\n",
    "iqr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_state.boxplot(column='Discharges');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will now create linear models to see if the population of a state inflences the medicare coverage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_state.rename(columns={'Avg % Covered':'Covered'}, inplace=True)\n",
    "\n",
    "pop_median = df_state.loc[:,'Population'].mean()\n",
    "dis_median = df_state.loc[:,'Discharges'].mean()\n",
    "\n",
    "def stand1(pop):\n",
    "    \n",
    "    if (pop <= pop_median):\n",
    "        return 'low'\n",
    "    else:\n",
    "        return 'high'\n",
    "    \n",
    "def stand2(dis):\n",
    "    \n",
    "    if (dis <= dis_median):\n",
    "        return 'low'\n",
    "    else:\n",
    "        return 'high'\n",
    "    \n",
    "df_state['Pop'] = df_state['Population'].apply(stand1)\n",
    "df_state['Dis'] = df_state['Discharges'].apply(stand2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The cell above calculates the mean populations and number of discharges. It then creates two new columns in df_state, which generalizes population and discharges into two categories based on their relation to the mean. Now we can create our OLS model to predict average percent covered from population."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                Covered   R-squared:                       0.069\n",
      "Model:                            OLS   Adj. R-squared:                  0.050\n",
      "Method:                 Least Squares   F-statistic:                     3.646\n",
      "Date:                Wed, 12 Jun 2019   Prob (F-statistic):             0.0621\n",
      "Time:                        03:49:14   Log-Likelihood:                 43.610\n",
      "No. Observations:                  51   AIC:                            -83.22\n",
      "Df Residuals:                      49   BIC:                            -79.36\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.2498      0.025      9.813      0.000       0.199       0.301\n",
      "Pop[T.low]     0.0595      0.031      1.909      0.062      -0.003       0.122\n",
      "==============================================================================\n",
      "Omnibus:                       59.420   Durbin-Watson:                   1.466\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              443.196\n",
      "Skew:                           2.972   Prob(JB):                     5.77e-97\n",
      "Kurtosis:                      16.162   Cond. No.                         3.23\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "outcome_1, predictors_1 = patsy.dmatrices('Covered ~ Pop',df_state)\n",
    "mod_1 = sm.OLS(outcome_1,predictors_1)\n",
    "res_1 = mod_1.fit()\n",
    "\n",
    "print(res_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the model above and using an alpha value of 0.01, population does not significantly predict average percent covered. We will create another OLS model to predict average percent covered from discharges."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                Covered   R-squared:                       0.002\n",
      "Model:                            OLS   Adj. R-squared:                 -0.018\n",
      "Method:                 Least Squares   F-statistic:                    0.1209\n",
      "Date:                Wed, 12 Jun 2019   Prob (F-statistic):              0.730\n",
      "Time:                        03:49:14   Log-Likelihood:                 41.843\n",
      "No. Observations:                  51   AIC:                            -79.69\n",
      "Df Residuals:                      49   BIC:                            -75.82\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "Intercept      0.2827      0.025     11.338      0.000       0.233       0.333\n",
      "Dis[T.low]     0.0109      0.031      0.348      0.730      -0.052       0.074\n",
      "==============================================================================\n",
      "Omnibus:                       61.128   Durbin-Watson:                   1.577\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              490.204\n",
      "Skew:                           3.054   Prob(JB):                    3.58e-107\n",
      "Kurtosis:                      16.906   Cond. No.                         3.04\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "outcome_2, predictors_2 = patsy.dmatrices('Covered ~ Dis',df_state)\n",
    "mod_2 = sm.OLS(outcome_2,predictors_2)\n",
    "res_2 = mod_2.fit()\n",
    "\n",
    "print(res_2.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the model above and using an alpha value of 0.01, discharges do not significantly predict average percent covered. We will create a few more linear models to analyze each condition separately. First we will apply the same changes from above to generalize population and discharges. To save space, we will only print out the linear coefficients relating Population to Covered (the slope estimate) and the P-values resulting from each model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "190 - CHRONIC OBSTRUCTIVE PULMONARY DISEASE W MCC\n",
      "\t Coefficient:  0.06906954524563157\n",
      "\t P-value:  0.046391265602517294\n",
      "194 - SIMPLE PNEUMONIA & PLEURISY W CC\n",
      "\t Coefficient:  0.06751140214684292\n",
      "\t P-value:  0.04404436370443962\n",
      "291 - HEART FAILURE & SHOCK W MCC\n",
      "\t Coefficient:  0.06194584190575887\n",
      "\t P-value:  0.07907914786449713\n",
      "292 - HEART FAILURE & SHOCK W CC\n",
      "\t Coefficient:  0.06635917661454908\n",
      "\t P-value:  0.047191210296787876\n",
      "392 - ESOPHAGITIS, GASTROENT & MISC DIGEST DISORDERS W/O MCC\n",
      "\t Coefficient:  0.05676146615886751\n",
      "\t P-value:  0.0622940600940216\n",
      "470 - MAJOR JOINT REPLACEMENT OR REATTACHMENT OF LOWER EXTREMITY W/O MCC\n",
      "\t Coefficient:  0.06738265368378277\n",
      "\t P-value:  0.03044779325792945\n",
      "641 - MISC DISORDERS OF NUTRITION,METABOLISM,FLUIDS/ELECTROLYTES W/O MCC\n",
      "\t Coefficient:  0.06081117820428253\n",
      "\t P-value:  0.043184867805011296\n",
      "683 - RENAL FAILURE W CC\n",
      "\t Coefficient:  0.06306959669948764\n",
      "\t P-value:  0.06777661068272371\n",
      "690 - KIDNEY & URINARY TRACT INFECTIONS W/O MCC\n",
      "\t Coefficient:  0.06953352031716535\n",
      "\t P-value:  0.030987444515706866\n",
      "871 - SEPTICEMIA OR SEVERE SEPSIS W/O MV 96+ HOURS W MCC\n",
      "\t Coefficient:  0.05389659656534554\n",
      "\t P-value:  0.11164480205875722\n"
     ]
    }
   ],
   "source": [
    "df_condition.rename(columns={'Avg % Covered':'Covered'}, inplace=True)\n",
    "\n",
    "df_condition['Pop'] = df_condition['Population'].apply(stand1)\n",
    "df_condition['Dis'] = df_condition['Discharges'].apply(stand2)\n",
    "\n",
    "for name in df_condition['DRG'].unique():\n",
    "    outcome, predictors = patsy.dmatrices('Covered ~ Pop',\n",
    "                    df_condition.loc[df_condition['DRG'] == name])\n",
    "    mod = sm.OLS(outcome, predictors)\n",
    "    res = mod.fit()\n",
    "    \n",
    "    print(name)\n",
    "    print('\\t Coefficient: ', res.params[1])\n",
    "    print('\\t P-value: ', res.f_pvalue)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There do not appear to be significant differences between any of the coefficients of the linear models. There are small differences in the P-values, but none are lower than 0.01, so we cannot conclude that population predicts average percent covered for any of these conditions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ethics & Privacy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The stakeholders in the hypothesis we are investigating include insurance companies and people who receive Medicare. One possible effect of our results could be that insurance companies could use the data to charge higher prices in regions with higher populations, which they would profit from. This might harm people with lower incomes who live in areas with high populations.\n",
    "\n",
    "The data we are using is collected from across over 3,000 U.S. hospitals that received Medicare payments, where the information is limited to the locations of the hospital and the provider and the amount covered. The data is likely not biased because data was collected from many hospitals in all 50 states, making it a comprehensive representation of all Medicare payments in the U.S. The patients’ data is kept anonymous. However, to further protect patients’ data, we drop all the columns in the data that we don’t need and only use the overall averages for each state and illness.\n",
    "\n",
    "The second dataset involves population, birth rates, death rates, and incomes in each state. Compared to the first dataset, it has less personal information. Hence, it would allow us to analyze the data collected by each state’s government. It doesn’t target any specific class, race or gender. Indeed, there are still potential biases: in this case, we can't guarantee that each state government would report accurate population data. The accuracy of the estimated population could be affected by several factors: such as undocumented people, human trafficking, etc.\n",
    "\n",
    "This may lead to potential biases because the dataset of the population can't and didn't include these marginalized groups of people into considerations. Hence, the dataset of population change may not fully reflect the real situation in each state. These issues are out of our capability to handle and can't be solved by us; therefore, we acknowledge these potential biases in the Ethics & Privacy section to present our thinking and concerns to these groups of people.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion & Discussion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We started off by cleaning the data and adding columns for the population of each state and average percent covered by Medicare, along with narrowing the data down to the top 10 conditions. We then performed data exploration by looking at summaries of the data and visualizations. For the data exploration, we analyzed correlation using heatmaps to obtain the different correlations between the different attributes of our data. We wanted to focus on the correlation between population and total payments and between population and average covered payments. Based on the heatmaps from our two datasets, we obtained a correlation of 0.52 between population and average covered charges and a correlation of 0.35 between population and average covered charges. These correlations aren’t significant enough to determine a strong relationship between population and average covered charges. To analyze our data, we performed an OLS regression to see if population and Medicare were correlated.\n",
    "\n",
    "Based on the results of the OLS regression, the P-values we got were not significant enough to give us a conclusive result about our analysis (assuming an alpha value of 0.01). The P-values we got from analyzing conditions separately were also not significant enough. Thus, we cannot conclude that the population of a state affects the average Medicare one receives.\n",
    "\n",
    "For the limitations of the analysis, we only have access to the public dataset. In other words, we didn’t include information and data from private companies. Hence, our dataset resource is not comprehensive enough to show the relation between the state population and medical insurance. For further research, we would like to include more dataset both from private companies and public platforms. Our project could also analyze the datasets and information on a global scale, reaching out to the international communities to see what determines the Medicare amount received by people. \n"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}