ronyarmon/reduction.ipynb

## reduction.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculate differences in antibiotic prescription\n",
    "Public Health England (PHE) datasets for Clinical Commissioning Groups (CCGs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "extensions": {
     "jupyter_dashboards": {
      "version": 1,
      "views": {
       "grid_default": {
        "hidden": true
       },
       "report_default": {
        "hidden": true
       }
      }
     }
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import re\n",
    "import datetime\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import plotly.plotly as py\n",
    "import cufflinks as cf\n",
    "import plotly.graph_objs as go"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "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>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "      <th>diff.</th>\n",
       "      <th>%diff.</th>\n",
       "      <th>AreaCode</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Wokingham</th>\n",
       "      <td>1.096361</td>\n",
       "      <td>0.128674</td>\n",
       "      <td>0.887632</td>\n",
       "      <td>0.984855</td>\n",
       "      <td>1.114881</td>\n",
       "      <td>1.224667</td>\n",
       "      <td>1.258441</td>\n",
       "      <td>0.370809</td>\n",
       "      <td>41.8</td>\n",
       "      <td>E38000209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Southend</th>\n",
       "      <td>1.312266</td>\n",
       "      <td>0.134439</td>\n",
       "      <td>1.110766</td>\n",
       "      <td>1.214596</td>\n",
       "      <td>1.299737</td>\n",
       "      <td>1.427133</td>\n",
       "      <td>1.526374</td>\n",
       "      <td>0.415608</td>\n",
       "      <td>37.4</td>\n",
       "      <td>E38000168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Greenwich</th>\n",
       "      <td>1.008254</td>\n",
       "      <td>0.086350</td>\n",
       "      <td>0.876274</td>\n",
       "      <td>0.948608</td>\n",
       "      <td>0.978116</td>\n",
       "      <td>1.085627</td>\n",
       "      <td>1.153810</td>\n",
       "      <td>0.277536</td>\n",
       "      <td>31.7</td>\n",
       "      <td>E38000066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enfield</th>\n",
       "      <td>0.980238</td>\n",
       "      <td>0.086953</td>\n",
       "      <td>0.866125</td>\n",
       "      <td>0.912279</td>\n",
       "      <td>0.943843</td>\n",
       "      <td>1.053164</td>\n",
       "      <td>1.139568</td>\n",
       "      <td>0.273442</td>\n",
       "      <td>31.6</td>\n",
       "      <td>E38000057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ealing</th>\n",
       "      <td>0.922385</td>\n",
       "      <td>0.092077</td>\n",
       "      <td>0.807871</td>\n",
       "      <td>0.851348</td>\n",
       "      <td>0.869793</td>\n",
       "      <td>1.029526</td>\n",
       "      <td>1.062063</td>\n",
       "      <td>0.254192</td>\n",
       "      <td>31.5</td>\n",
       "      <td>E38000048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Southwark</th>\n",
       "      <td>0.822577</td>\n",
       "      <td>0.070574</td>\n",
       "      <td>0.711615</td>\n",
       "      <td>0.772357</td>\n",
       "      <td>0.799748</td>\n",
       "      <td>0.892505</td>\n",
       "      <td>0.934390</td>\n",
       "      <td>0.222775</td>\n",
       "      <td>31.3</td>\n",
       "      <td>E38000171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vale Royal</th>\n",
       "      <td>1.124266</td>\n",
       "      <td>0.091501</td>\n",
       "      <td>0.954416</td>\n",
       "      <td>1.080340</td>\n",
       "      <td>1.101124</td>\n",
       "      <td>1.220634</td>\n",
       "      <td>1.247929</td>\n",
       "      <td>0.293513</td>\n",
       "      <td>30.8</td>\n",
       "      <td>E38000189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Corby</th>\n",
       "      <td>1.088365</td>\n",
       "      <td>0.085133</td>\n",
       "      <td>0.999059</td>\n",
       "      <td>1.033148</td>\n",
       "      <td>1.052709</td>\n",
       "      <td>1.125940</td>\n",
       "      <td>1.300060</td>\n",
       "      <td>0.301001</td>\n",
       "      <td>30.1</td>\n",
       "      <td>E38000037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>West Lancashire</th>\n",
       "      <td>1.128477</td>\n",
       "      <td>0.083207</td>\n",
       "      <td>0.955026</td>\n",
       "      <td>1.082946</td>\n",
       "      <td>1.116282</td>\n",
       "      <td>1.213895</td>\n",
       "      <td>1.236643</td>\n",
       "      <td>0.281617</td>\n",
       "      <td>29.5</td>\n",
       "      <td>E38000200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Barking And Dagenham</th>\n",
       "      <td>1.098038</td>\n",
       "      <td>0.092103</td>\n",
       "      <td>0.960762</td>\n",
       "      <td>1.029688</td>\n",
       "      <td>1.059582</td>\n",
       "      <td>1.202810</td>\n",
       "      <td>1.241309</td>\n",
       "      <td>0.280547</td>\n",
       "      <td>29.2</td>\n",
       "      <td>E38000004</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          mean       std       min       25%       50%  \\\n",
       "Wokingham             1.096361  0.128674  0.887632  0.984855  1.114881   \n",
       "Southend              1.312266  0.134439  1.110766  1.214596  1.299737   \n",
       "Greenwich             1.008254  0.086350  0.876274  0.948608  0.978116   \n",
       "Enfield               0.980238  0.086953  0.866125  0.912279  0.943843   \n",
       "Ealing                0.922385  0.092077  0.807871  0.851348  0.869793   \n",
       "Southwark             0.822577  0.070574  0.711615  0.772357  0.799748   \n",
       "Vale Royal            1.124266  0.091501  0.954416  1.080340  1.101124   \n",
       "Corby                 1.088365  0.085133  0.999059  1.033148  1.052709   \n",
       "West Lancashire       1.128477  0.083207  0.955026  1.082946  1.116282   \n",
       "Barking And Dagenham  1.098038  0.092103  0.960762  1.029688  1.059582   \n",
       "\n",
       "                           75%       max     diff.  %diff.   AreaCode  \n",
       "Wokingham             1.224667  1.258441  0.370809    41.8  E38000209  \n",
       "Southend              1.427133  1.526374  0.415608    37.4  E38000168  \n",
       "Greenwich             1.085627  1.153810  0.277536    31.7  E38000066  \n",
       "Enfield               1.053164  1.139568  0.273442    31.6  E38000057  \n",
       "Ealing                1.029526  1.062063  0.254192    31.5  E38000048  \n",
       "Southwark             0.892505  0.934390  0.222775    31.3  E38000171  \n",
       "Vale Royal            1.220634  1.247929  0.293513    30.8  E38000189  \n",
       "Corby                 1.125940  1.300060  0.301001    30.1  E38000037  \n",
       "West Lancashire       1.213895  1.236643  0.281617    29.5  E38000200  \n",
       "Barking And Dagenham  1.202810  1.241309  0.280547    29.2  E38000004  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#load datasets\n",
    "ccgs_92377=pd.read_csv('ccgs_92377_clean.csv',low_memory=False)\n",
    "timeperiods=ccgs_92377['Timeperiod'].unique()\n",
    "\n",
    "#list of ccg names \n",
    "ccgs_list=ccgs_92377['CCG'].unique()\n",
    "ccgs_list=ccgs_list.tolist()\n",
    "\n",
    "#establish time series per individual ccgs as columns\n",
    "all_ccgs=pd.DataFrame()\n",
    "for index, value in enumerate (ccgs_list):\n",
    "        indicator_ccg=ccgs_92377[ccgs_92377['CCG']==value]\n",
    "        indicator_ccg_values=indicator_ccg.loc[:,'Value']\n",
    "        all_ccgs [value]=pd.Series()\n",
    "        all_ccgs [value]=indicator_ccg_values.values #store the indicator data for all ccgs\n",
    "\n",
    "# store descriptive stats for each CCG\n",
    "descriptive_all=pd.DataFrame(all_ccgs.describe())\n",
    "descriptive_all=descriptive_all.transpose() #ccgs as indexed rows and stats as columns\n",
    "del descriptive_all['count']\n",
    "\n",
    "#calculate max-min difference\n",
    "descriptive_all['diff.']=descriptive_all['max']-descriptive_all['min']\n",
    "descriptive_all['%diff.']=round(100*(descriptive_all['diff.']/descriptive_all['min']),1)\n",
    "descriptive_all['AreaCode']=area_codes=ccgs_92377['AreaCode'].unique()\n",
    "\n",
    "#sort by difference values\n",
    "descriptive_all_sorted=descriptive_all.sort_values(by=['%diff.'],ascending=False)\n",
    "descriptive_all_sorted.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a188c1898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(font_scale = 1.3)\n",
    "\n",
    "plt.subplot(1,2, 1)\n",
    "kx=descriptive_all_sorted['%diff.'].plot(color=['green'], figsize=(12,6))\n",
    "plt.xlabel('CCG list (Sorted by Difference)',fontsize=14)\n",
    "plt.ylabel('% Difference', fontsize=14)\n",
    "kx.set_title('Reduction per CCG\\n', fontsize=18)\n",
    "kx.margins(0.02)\n",
    "\n",
    "plt.subplot(1,2, 2)\n",
    "lx=descriptive_all_sorted['%diff.'].plot(kind='hist',bins=20, color=['green'], figsize=(12,6))\n",
    "lx.set_title('Reduction frequencies\\n', fontsize=18)\n",
    "plt.xlabel('% Difference',fontsize=14)\n",
    "plt.ylabel('Number of CCGs', fontsize=14)\n",
    "lx.margins(0.02)\n",
    "\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<iframe id=\"igraph\" scrolling=\"no\" style=\"border:none;\" seamless=\"seamless\" src=\"https://plot.ly/~rony_armon/89.embed\" height=\"525px\" width=\"100%\"></iframe>"
      ],
      "text/plain": [
       "<plotly.tools.PlotlyDisplay object>"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#use plotly to show individual ccgs\n",
    "data = [\n",
    "    go.Scatter(\n",
    "        x=descriptive_all.index, \n",
    "        y=descriptive_all_sorted['%diff.'],\n",
    "        mode = 'lines+markers',\n",
    "        line = dict(\n",
    "        color = ('rgb(40, 140, 60)'),\n",
    "        width = 2)\n",
    "    )\n",
    "]\n",
    "py.iplot(data, filename='%diff_ccgs')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "extensions": {
     "jupyter_dashboards": {
      "version": 1,
      "views": {
       "grid_default": {
        "col": 7,
        "height": 4,
        "hidden": true,
        "row": 17,
        "width": 4
       },
       "report_default": {
        "hidden": false
       }
      }
     }
    }
   },
   "source": [
    "Findings: The typical difference between minimal and maximal levels of perscription is aruond 10%. \n",
    "Some trusts however, showed much higher levels of difference, between 25-40%. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "#results to csvs for dashboard\n",
    "descriptive_all.to_csv('diff_2014_2017.csv')"
   ]
  }
 ],
 "metadata": {
  "extensions": {
   "jupyter_dashboards": {
    "activeView": "grid_default",
    "version": 1,
    "views": {
     "grid_default": {
      "cellMargin": 10,
      "defaultCellHeight": 20,
      "maxColumns": 12,
      "name": "grid",
      "type": "grid"
     },
     "report_default": {
      "name": "report",
      "type": "report"
     }
    }
   }
  },
  "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}