{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MIMIC-II IAC Causal Inference"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This tutorial continues the exploration the MIMIC-II IAC dataset using causal inference methods. The dataset was created for the purpose of a case study in the book: Secondary Analysis of Electronic Health Records, published by Springer in 2016. In particular, the dataset was used throughout Chapter 16 (Data Analysis) by Raffa J. et al. to investigate the effectiveness of indwelling arterial catheters in hemodynamically stable patients with respiratory failure for mortality outcomes. The dataset is derived from MIMIC-II, the publicly-accessible critical care database. It contains summary clinical data and  outcomes for 1,776 patients.\n",
    "\n",
    "Reference: \n",
    "\n",
    "[1] https://github.com/py-why/dowhy\n",
    "\n",
    "[2] https://www.pywhy.org/dowhy/\n",
    "\n",
    "[3] https://arxiv.org/abs/2011.04216\n",
    "\n",
    "[4] https://physionet.org/content/mimic2-iaccd/1.0/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing ehrapy and setting plotting parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import ehrapy as ep\n",
    "from IPython.display import display\n",
    "import graphviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----\n",
      "ehrapy              0.9.0\n",
      "rich                NA\n",
      "scanpy              1.9.3\n",
      "session_info        1.0.0\n",
      "-----\n",
      "CoreFoundation              NA\n",
      "Foundation                  NA\n",
      "Levenshtein                 0.21.1\n",
      "PIL                         10.0.0\n",
      "PyObjCTools                 NA\n",
      "anndata                     0.9.2\n",
      "anyio                       NA\n",
      "appnope                     0.1.3\n",
      "argcomplete                 NA\n",
      "arrow                       1.2.3\n",
      "astor                       0.8.1\n",
      "asttokens                   NA\n",
      "attr                        23.1.0\n",
      "attrs                       23.1.0\n",
      "autograd                    NA\n",
      "autograd_gamma              NA\n",
      "babel                       2.12.1\n",
      "backcall                    0.2.0\n",
      "brotli                      1.0.9\n",
      "cachetools                  5.3.1\n",
      "causallearn                 NA\n",
      "certifi                     2023.07.22\n",
      "cffi                        1.15.1\n",
      "charset_normalizer          3.2.0\n",
      "cloudpickle                 2.2.1\n",
      "comm                        0.1.3\n",
      "cvxopt                      1.3.1\n",
      "cycler                      0.10.0\n",
      "cython_runtime              NA\n",
      "dateutil                    2.8.2\n",
      "db_dtypes                   1.1.1\n",
      "debugpy                     1.6.7\n",
      "decorator                   5.1.1\n",
      "defusedxml                  0.7.1\n",
      "dill                        0.3.7\n",
      "dot_parser                  NA\n",
      "dowhy                       0.11.1\n",
      "executing                   1.2.0\n",
      "fastjsonschema              NA\n",
      "fhiry                       3.0.0\n",
      "filelock                    3.12.2\n",
      "formulaic                   0.6.4\n",
      "fqdn                        NA\n",
      "future                      0.18.3\n",
      "gmpy2                       2.1.2\n",
      "google                      NA\n",
      "graphlib                    NA\n",
      "graphviz                    0.20.1\n",
      "grpc                        1.56.2\n",
      "grpc_status                 NA\n",
      "h5py                        3.9.0\n",
      "idna                        3.4\n",
      "igraph                      0.10.6\n",
      "imblearn                    0.12.3\n",
      "interface_meta              1.3.0\n",
      "ipykernel                   6.25.0\n",
      "ipywidgets                  8.0.7\n",
      "isoduration                 NA\n",
      "jedi                        0.19.0\n",
      "jinja2                      3.1.4\n",
      "joblib                      1.3.0\n",
      "json5                       NA\n",
      "jsonpointer                 2.0\n",
      "jsonschema                  4.18.4\n",
      "jsonschema_specifications   NA\n",
      "jupyter_events              0.7.0\n",
      "jupyter_server              2.7.0\n",
      "jupyterlab_server           2.24.0\n",
      "kiwisolver                  1.4.4\n",
      "lamin_utils                 0.13.2\n",
      "leidenalg                   0.10.1\n",
      "lifelines                   0.27.7\n",
      "llvmlite                    0.40.1\n",
      "markupsafe                  2.1.3\n",
      "matplotlib                  3.7.2\n",
      "missingno                   0.5.2\n",
      "mpl_toolkits                NA\n",
      "mpmath                      1.3.0\n",
      "natsort                     8.4.0\n",
      "nbformat                    5.9.2\n",
      "networkx                    3.1\n",
      "numba                       0.57.1\n",
      "numpy                       1.24.0\n",
      "objc                        9.2\n",
      "overrides                   NA\n",
      "packaging                   23.1\n",
      "pandas                      2.0.3\n",
      "parso                       0.8.3\n",
      "patsy                       0.5.6\n",
      "pexpect                     4.8.0\n",
      "pickleshare                 0.7.5\n",
      "pkg_resources               NA\n",
      "platformdirs                3.10.0\n",
      "prometheus_client           NA\n",
      "prompt_toolkit              3.0.39\n",
      "psutil                      5.9.5\n",
      "ptyprocess                  0.7.0\n",
      "pure_eval                   0.2.2\n",
      "pyarrow                     12.0.1\n",
      "pyasn1                      0.5.0\n",
      "pyasn1_modules              0.3.0\n",
      "pydev_ipython               NA\n",
      "pydevconsole                NA\n",
      "pydevd                      2.9.5\n",
      "pydevd_file_utils           NA\n",
      "pydevd_plugins              NA\n",
      "pydevd_tracing              NA\n",
      "pydot                       1.4.2\n",
      "pygments                    2.15.1\n",
      "pyparsing                   3.0.9\n",
      "pythonjsonlogger            NA\n",
      "pytz                        2023.3\n",
      "rapidfuzz                   3.1.2\n",
      "referencing                 NA\n",
      "requests                    2.31.0\n",
      "rfc3339_validator           0.1.4\n",
      "rfc3986_validator           0.1.1\n",
      "rpds                        NA\n",
      "rsa                         4.9\n",
      "scipy                       1.11.1\n",
      "seaborn                     0.12.2\n",
      "send2trash                  NA\n",
      "setuptools                  68.0.0\n",
      "six                         1.16.0\n",
      "sklearn                     1.3.0\n",
      "sniffio                     1.3.0\n",
      "socks                       1.7.1\n",
      "sphinxcontrib               NA\n",
      "stack_data                  0.6.2\n",
      "statsmodels                 0.14.2\n",
      "sympy                       1.12\n",
      "tableone                    0.9.1\n",
      "tabulate                    0.9.0\n",
      "texttable                   1.6.7\n",
      "thefuzz                     0.19.0\n",
      "threadpoolctl               3.2.0\n",
      "torch                       2.0.1\n",
      "tornado                     6.3.2\n",
      "tqdm                        4.65.0\n",
      "traitlets                   5.9.0\n",
      "typing_extensions           NA\n",
      "uri_template                NA\n",
      "urllib3                     2.0.4\n",
      "vscode                      NA\n",
      "wcwidth                     0.2.6\n",
      "webcolors                   1.13\n",
      "websocket                   1.6.1\n",
      "wrapt                       1.15.0\n",
      "yaml                        6.0\n",
      "zmq                         25.1.0\n",
      "zoneinfo                    NA\n",
      "-----\n",
      "IPython             8.14.0\n",
      "jupyter_client      8.3.0\n",
      "jupyter_core        5.3.1\n",
      "jupyterlab          4.0.3\n",
      "notebook            7.0.1\n",
      "-----\n",
      "Python 3.10.12 | packaged by conda-forge | (main, Jun 23 2023, 22:41:52) [Clang 15.0.7 ]\n",
      "macOS-13.1-arm64-arm-64bit\n",
      "-----\n",
      "Session information updated at 2024-06-28 11:35\n"
     ]
    }
   ],
   "source": [
    "ep.print_versions()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MIMIC-II dataset preparation\n",
    "Let's load the MIMIC-II dataset using ehrapy with default one-hot encoding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "637ed865f5ab40b8a969ed6f5327d844",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "adata = ep.dt.mimic_2(encoded=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The MIMIC-II dataset has 1776 patients as described above with 46 features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 1776 × 54\n",
       "    obs: 'service_unit', 'day_icu_intime'\n",
       "    var: 'feature_type', 'unencoded_var_names', 'encoding_mode'\n",
       "    layers: 'original'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Causal Inference on the MIMIC-II dataset\n",
    "In the background, `ehrapy` uses the `dowhy` package to enable effortless causal inference on electronic health records (EHR). Any `dowhy` analysis is structured into 3 steps:\n",
    "1) Formulate causal questions\n",
    "2) Estimate causal effects\n",
    "3) Perform refutation tests.\n",
    "\n",
    "Within `ehrapy`, we have consolidated this method into a single function `ehrapy.causal_inference()`. This function takes in the dataset, the treatment variable, the outcome variable, and the further optional parameters. It then performs the 3 steps above and displays the information in a user-friendly way."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Causal Graph\n",
    "\n",
    "The causal graph is a directed acyclic graph (DAG) that represents the causal relationships between the variables in the dataset. Here, we create it by manually writing out the connections between the variables. Other options would be the GML or DOT graph format. Furthermore, you can use graphical tools like [DAGitty](http://dagitty.net/dags.html) to construct the graph. You can export the graph string that it generates. The graph string is very close to the DOT format: just rename dag to digraph, remove newlines and add a semicolon after every line, to convert it to the DOT format and input to DoWhy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Assumptions:\n",
    " - both age and overweight increase your risk for medical problems\n",
    " - having a lot of problems makes you more likely to die in the hospital\n",
    " - having a lot of problems influences your likelihood of getting an IAC\n",
    " - having an IAC influences your likelihood of dying in the hospital"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 8.1.0 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"919pt\" height=\"260pt\"\n",
       " viewBox=\"0.00 0.00 918.88 260.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 256)\">\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-256 914.88,-256 914.88,4 -4,4\"/>\n",
       "<!-- aline_flg -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>aline_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"432.77\" cy=\"-90\" rx=\"134.26\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"432.77\" y=\"-84.95\" font-family=\"Times,serif\" font-size=\"14.00\">Indwelling arterial catheters used</text>\n",
       "</g>\n",
       "<!-- icu_los_day -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>icu_los_day</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"513.77\" cy=\"-18\" rx=\"58\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"513.77\" y=\"-12.95\" font-family=\"Times,serif\" font-size=\"14.00\">Days in ICU</text>\n",
       "</g>\n",
       "<!-- aline_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge37\" class=\"edge\">\n",
       "<title>aline_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M452.38,-72.05C462.67,-63.16 475.45,-52.11 486.68,-42.41\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"488.62,-44.49 493.9,-35.31 484.04,-39.2 488.62,-44.49\"/>\n",
       "</g>\n",
       "<!-- age -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>age</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"417.77\" cy=\"-234\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"417.77\" y=\"-228.95\" font-family=\"Times,serif\" font-size=\"14.00\">age</text>\n",
       "</g>\n",
       "<!-- sepsis_flg -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>sepsis_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"47.77\" cy=\"-162\" rx=\"47.77\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"47.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">sepsis_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;sepsis_flg -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>age&#45;&gt;sepsis_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M391.07,-229.66C336.62,-222.6 208.98,-204.74 103.77,-180 100.41,-179.21 96.95,-178.33 93.48,-177.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"94.71,-173.82 84.13,-174.48 92.81,-180.56 94.71,-173.82\"/>\n",
       "</g>\n",
       "<!-- chf_flg -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>chf_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"150.77\" cy=\"-162\" rx=\"37.53\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"150.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">chf_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;chf_flg -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>age&#45;&gt;chf_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M392.12,-227.79C350.82,-219.17 267.19,-200.85 197.77,-180 195.15,-179.21 192.47,-178.36 189.77,-177.48\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"191.03,-173.87 180.43,-173.94 188.76,-180.49 191.03,-173.87\"/>\n",
       "</g>\n",
       "<!-- afib_flg -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>afib_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"246.77\" cy=\"-162\" rx=\"40.09\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"246.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">afib_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;afib_flg -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>age&#45;&gt;afib_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M395.06,-223.7C367.15,-212.28 319.34,-192.71 285.57,-178.88\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"287.15,-175.34 276.57,-174.79 284.5,-181.82 287.15,-175.34\"/>\n",
       "</g>\n",
       "<!-- renal_flg -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>renal_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"348.77\" cy=\"-162\" rx=\"44.19\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"348.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">renal_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;renal_flg -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>age&#45;&gt;renal_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M403.47,-218.5C394.48,-209.37 382.65,-197.37 372.3,-186.87\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"375.25,-184.86 365.74,-180.2 370.26,-189.78 375.25,-184.86\"/>\n",
       "</g>\n",
       "<!-- liver_flg -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>liver_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"453.77\" cy=\"-162\" rx=\"42.65\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"453.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">liver_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;liver_flg -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>age&#45;&gt;liver_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M426.12,-216.76C430.33,-208.58 435.54,-198.45 440.3,-189.2\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"443.75,-191.13 445.21,-180.63 437.53,-187.93 443.75,-191.13\"/>\n",
       "</g>\n",
       "<!-- copd_flg -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>copd_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"557.77\" cy=\"-162\" rx=\"43.67\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"557.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">copd_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;copd_flg -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>age&#45;&gt;copd_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M439.03,-222.37C460.87,-211.45 495.28,-194.24 521.44,-181.16\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"522.65,-183.97 530.03,-176.37 519.52,-177.71 522.65,-183.97\"/>\n",
       "</g>\n",
       "<!-- cad_flg -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>cad_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"657.77\" cy=\"-162\" rx=\"38.56\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"657.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">cad_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;cad_flg -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>age&#45;&gt;cad_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M438.77,-222.15C443.62,-219.91 448.8,-217.72 453.77,-216 521.42,-192.58 542.09,-200.21 610.77,-180 613.23,-179.28 615.74,-178.49 618.27,-177.67\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"619.27,-180.69 627.62,-174.17 617.02,-174.06 619.27,-180.69\"/>\n",
       "</g>\n",
       "<!-- stroke_flg -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>stroke_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"762.77\" cy=\"-162\" rx=\"48.28\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"762.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">stroke_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;stroke_flg -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>age&#45;&gt;stroke_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M438.29,-222.02C443.25,-219.73 448.61,-217.55 453.77,-216 562.13,-183.48 595.26,-204.23 705.77,-180 709.36,-179.21 713.05,-178.32 716.75,-177.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"717.38,-180.54 726.1,-174.52 715.53,-173.79 717.38,-180.54\"/>\n",
       "</g>\n",
       "<!-- resp_flg -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>resp_flg</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"869.77\" cy=\"-162\" rx=\"41.12\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"869.77\" y=\"-156.95\" font-family=\"Times,serif\" font-size=\"14.00\">resp_flg</text>\n",
       "</g>\n",
       "<!-- age&#45;&gt;resp_flg -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>age&#45;&gt;resp_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M438.24,-221.86C443.2,-219.58 448.57,-217.44 453.77,-216 611.32,-172.46 660.01,-214.54 819.77,-180 822.79,-179.35 825.89,-178.57 828.99,-177.71\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"829.85,-180.81 838.41,-174.56 827.82,-174.11 829.85,-180.81\"/>\n",
       "</g>\n",
       "<!-- sepsis_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge19\" class=\"edge\">\n",
       "<title>sepsis_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M83.12,-149.48C89.93,-147.49 97.04,-145.56 103.77,-144 177.45,-126.96 261.67,-113.51 326.45,-104.41\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"326.66,-107.78 336.09,-102.94 325.7,-100.85 326.66,-107.78\"/>\n",
       "</g>\n",
       "<!-- sepsis_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge28\" class=\"edge\">\n",
       "<title>sepsis_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M68.53,-145.39C96.55,-125.13 148.74,-90.24 198.77,-72 280.77,-42.1 381.01,-28.93 446.07,-23.22\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"446.32,-26.63 455.99,-22.31 445.74,-19.66 446.32,-26.63\"/>\n",
       "</g>\n",
       "<!-- chf_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge20\" class=\"edge\">\n",
       "<title>chf_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M179.41,-149.99C185.43,-147.86 191.77,-145.75 197.77,-144 246.82,-129.66 302.6,-116.93 347.49,-107.55\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"347.99,-110.81 357.07,-105.35 346.57,-103.96 347.99,-110.81\"/>\n",
       "</g>\n",
       "<!-- chf_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge29\" class=\"edge\">\n",
       "<title>chf_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M164.34,-145.04C182.54,-124.74 216.86,-90.14 253.77,-72 314.78,-42.01 391.86,-28.98 446.2,-23.33\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"446.48,-26.72 456.09,-22.26 445.8,-19.75 446.48,-26.72\"/>\n",
       "</g>\n",
       "<!-- afib_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge21\" class=\"edge\">\n",
       "<title>afib_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M276.65,-149.75C304.2,-139.39 345.74,-123.75 379.04,-111.22\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"379.96,-114.24 388.09,-107.44 377.49,-107.69 379.96,-114.24\"/>\n",
       "</g>\n",
       "<!-- afib_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge30\" class=\"edge\">\n",
       "<title>afib_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M248.3,-143.86C251.06,-123.65 258.69,-90.46 279.77,-72 304.71,-50.16 389.12,-34.99 449.39,-26.63\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"449.53,-30.01 458.97,-25.2 448.59,-23.08 449.53,-30.01\"/>\n",
       "</g>\n",
       "<!-- renal_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge22\" class=\"edge\">\n",
       "<title>renal_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M367.41,-145.46C378.07,-136.58 391.67,-125.25 403.7,-115.23\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"405.67,-117.3 411.12,-108.21 401.19,-111.92 405.67,-117.3\"/>\n",
       "</g>\n",
       "<!-- renal_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge31\" class=\"edge\">\n",
       "<title>renal_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M326.65,-145.98C302.65,-127.85 270.22,-96.72 289.77,-72 309.26,-47.36 389.29,-33 448.15,-25.58\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"448.37,-28.96 457.88,-24.27 447.53,-22.01 448.37,-28.96\"/>\n",
       "</g>\n",
       "<!-- liver_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge23\" class=\"edge\">\n",
       "<title>liver_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M448.58,-143.7C446.29,-136.07 443.54,-126.92 440.99,-118.4\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"444.13,-117.68 437.9,-109.1 437.42,-119.69 444.13,-117.68\"/>\n",
       "</g>\n",
       "<!-- liver_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge32\" class=\"edge\">\n",
       "<title>liver_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M485.22,-149.39C519.57,-136.4 570.45,-116.11 575.77,-108 584.54,-94.62 582.46,-86.53 575.77,-72 570.07,-59.64 559.92,-49.09 549.45,-40.74\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"551.8,-37.38 541.68,-34.24 547.64,-43.01 551.8,-37.38\"/>\n",
       "</g>\n",
       "<!-- copd_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge24\" class=\"edge\">\n",
       "<title>copd_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M532.47,-146.83C515.18,-137.15 491.92,-124.13 472.22,-113.09\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"474.31,-109.69 463.87,-107.86 470.89,-115.8 474.31,-109.69\"/>\n",
       "</g>\n",
       "<!-- copd_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge33\" class=\"edge\">\n",
       "<title>copd_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M578.69,-145.87C589.91,-136.41 602.66,-123.16 608.77,-108 614.75,-93.16 616.87,-85.8 608.77,-72 599.02,-55.39 581.96,-43.6 564.96,-35.43\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"566.58,-31.88 556.02,-31.03 563.74,-38.28 566.58,-31.88\"/>\n",
       "</g>\n",
       "<!-- cad_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge25\" class=\"edge\">\n",
       "<title>cad_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M628.27,-150.04C622.49,-147.97 616.46,-145.87 610.77,-144 574,-131.89 532.56,-119.48 498.92,-109.71\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"500.32,-106.19 489.75,-106.77 498.38,-112.91 500.32,-106.19\"/>\n",
       "</g>\n",
       "<!-- cad_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge34\" class=\"edge\">\n",
       "<title>cad_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M655.54,-143.57C652.2,-123.99 644.2,-92.31 625.77,-72 610.12,-54.76 587.69,-42.63 567.01,-34.36\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"568.31,-30.73 557.71,-30.5 565.85,-37.29 568.31,-30.73\"/>\n",
       "</g>\n",
       "<!-- stroke_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge26\" class=\"edge\">\n",
       "<title>stroke_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M726.73,-149.67C719.81,-147.66 712.59,-145.67 705.77,-144 646.11,-129.35 578.32,-116.17 525.21,-106.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"525.91,-103.01 515.45,-104.7 524.68,-109.9 525.91,-103.01\"/>\n",
       "</g>\n",
       "<!-- stroke_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge35\" class=\"edge\">\n",
       "<title>stroke_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M745.75,-144.72C724.77,-125.35 687.4,-92.99 650.77,-72 623.71,-56.5 591.25,-43.71 564.72,-34.59\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"566.08,-31.02 555.49,-31.15 563.85,-37.65 566.08,-31.02\"/>\n",
       "</g>\n",
       "<!-- resp_flg&#45;&gt;aline_flg -->\n",
       "<g id=\"edge27\" class=\"edge\">\n",
       "<title>resp_flg&#45;&gt;aline_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M839.1,-149.54C832.77,-147.47 826.1,-145.5 819.77,-144 731.99,-123.27 630.97,-109.77 554.15,-101.59\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"554.79,-98.03 544.48,-100.47 554.06,-105 554.79,-98.03\"/>\n",
       "</g>\n",
       "<!-- resp_flg&#45;&gt;icu_los_day -->\n",
       "<g id=\"edge36\" class=\"edge\">\n",
       "<title>resp_flg&#45;&gt;icu_los_day</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M845.19,-147.09C811.24,-128.28 747.74,-94.5 690.77,-72 650.75,-56.19 604.08,-42.4 568.71,-32.85\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"569.68,-29.22 559.12,-30.03 567.88,-35.99 569.68,-29.22\"/>\n",
       "</g>\n",
       "<!-- bmi -->\n",
       "<g id=\"node13\" class=\"node\">\n",
       "<title>bmi</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"489.77\" cy=\"-234\" rx=\"27\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"489.77\" y=\"-228.95\" font-family=\"Times,serif\" font-size=\"14.00\">bmi</text>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;sepsis_flg -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;sepsis_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M469.29,-221.87C464.33,-219.59 458.96,-217.45 453.77,-216 303.13,-174.01 257.07,-210.85 103.77,-180 100.17,-179.28 96.48,-178.42 92.78,-177.48\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"94.02,-173.92 83.46,-174.66 92.18,-180.67 94.02,-173.92\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;chf_flg -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;chf_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M469.25,-222.01C464.29,-219.73 458.93,-217.54 453.77,-216 343.68,-183.1 309.28,-207.69 197.77,-180 194.94,-179.3 192.05,-178.49 189.16,-177.61\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"190.4,-174.01 179.81,-174.21 188.22,-180.66 190.4,-174.01\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;afib_flg -->\n",
       "<g id=\"edge12\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;afib_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M468.77,-222.15C463.92,-219.91 458.73,-217.72 453.77,-216 385.69,-192.48 365.01,-199.83 295.77,-180 293.02,-179.21 290.21,-178.36 287.38,-177.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"288.78,-173.92 278.19,-174.13 286.6,-180.57 288.78,-173.92\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;renal_flg -->\n",
       "<g id=\"edge13\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;renal_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M468.66,-222.52C446.79,-211.66 412.19,-194.48 385.76,-181.37\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"387.59,-177.87 377.08,-176.56 384.48,-184.14 387.59,-177.87\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;liver_flg -->\n",
       "<g id=\"edge14\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;liver_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M481.42,-216.76C477.21,-208.58 472,-198.45 467.24,-189.2\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"470.01,-187.93 462.32,-180.63 463.78,-191.13 470.01,-187.93\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;copd_flg -->\n",
       "<g id=\"edge15\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;copd_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M503.86,-218.5C512.72,-209.37 524.38,-197.37 534.58,-186.87\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"536.58,-189.81 541.03,-180.2 531.56,-184.93 536.58,-189.81\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;cad_flg -->\n",
       "<g id=\"edge16\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;cad_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M512.43,-223.56C539.94,-212.09 586.77,-192.58 619.82,-178.81\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"620.72,-181.81 628.6,-174.74 618.02,-175.35 620.72,-181.81\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;stroke_flg -->\n",
       "<g id=\"edge17\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;stroke_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M515.15,-227.2C555.62,-217.86 637.21,-198.62 705.77,-180 709.11,-179.09 712.55,-178.13 716.01,-177.15\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"716.72,-180.29 725.36,-174.15 714.78,-173.57 716.72,-180.29\"/>\n",
       "</g>\n",
       "<!-- bmi&#45;&gt;resp_flg -->\n",
       "<g id=\"edge18\" class=\"edge\">\n",
       "<title>bmi&#45;&gt;resp_flg</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M516.6,-230.24C573.21,-224.09 708.83,-207.56 819.77,-180 822.73,-179.26 825.77,-178.43 828.81,-177.54\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"829.57,-180.66 838.07,-174.34 827.48,-173.98 829.57,-180.66\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.sources.Source at 0x2b65e7430>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "causal_graph = \"\"\"digraph {\n",
    "aline_flg[label=\"Indwelling arterial catheters used\"];\n",
    "icu_los_day[label=\"Days in ICU\"];\n",
    "\n",
    "age -> sepsis_flg;\n",
    "age -> chf_flg;\n",
    "age -> afib_flg;\n",
    "age -> renal_flg;\n",
    "age -> liver_flg;\n",
    "age -> copd_flg;\n",
    "age -> cad_flg;\n",
    "age -> stroke_flg;\n",
    "age -> resp_flg;\n",
    "bmi -> sepsis_flg;\n",
    "bmi -> chf_flg;\n",
    "bmi -> afib_flg;\n",
    "bmi -> renal_flg;\n",
    "bmi -> liver_flg;\n",
    "bmi -> copd_flg;\n",
    "bmi -> cad_flg;\n",
    "bmi -> stroke_flg;\n",
    "bmi -> resp_flg;\n",
    "sepsis_flg -> aline_flg;\n",
    "chf_flg -> aline_flg;\n",
    "afib_flg -> aline_flg;\n",
    "renal_flg -> aline_flg;\n",
    "liver_flg -> aline_flg;\n",
    "copd_flg -> aline_flg;\n",
    "cad_flg -> aline_flg;\n",
    "stroke_flg -> aline_flg;\n",
    "resp_flg -> aline_flg;\n",
    "sepsis_flg -> icu_los_day;\n",
    "chf_flg -> icu_los_day;\n",
    "afib_flg -> icu_los_day;\n",
    "renal_flg -> icu_los_day;\n",
    "liver_flg -> icu_los_day;\n",
    "copd_flg -> icu_los_day;\n",
    "cad_flg -> icu_los_day;\n",
    "stroke_flg -> icu_los_day;\n",
    "resp_flg -> icu_los_day;\n",
    "aline_flg -> icu_los_day;\n",
    "}\"\"\"\n",
    "\n",
    "g = graphviz.Source(causal_graph)\n",
    "display(g)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Causal inference"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This graph can now be fed into the `causal_inference()` method. Furthermore, we have to specify an `estimation_method`. For now, we will use `backdoor.linear_regression`.\n",
    "Please refer to [this example notebook](https://github.com/py-why/dowhy/blob/main/docs/source/example_notebooks/dowhy_estimation_methods.ipynb) and the [official dowhy documentation](https://www.pywhy.org/dowhy/) for more information on the different estimation methods. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (1/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (2/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (3/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (4/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (5/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (6/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (7/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (8/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (9/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (10/10)\n",
      "Causal inference results for treatment variable 'aline_flg' and outcome variable 'icu_los_day':\n",
      "└- Increasing the treatment variable(s) [aline_flg] from 0 to 1 causes an increase of 2.2348813091683186 in the expected value of the outcome [['icu_los_day']], over the data distribution/population represented by the dataset.\n",
      "\n",
      "Refutation results\n",
      "├-Refute: Use a Placebo Treatment\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 0.000\n",
      "|    ├- p-value: nan\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Add a random common cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.235\n",
      "|    ├- p-value: 0.494\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Use a subset of data\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.233\n",
      "|    ├- p-value: 0.489\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Add an Unobserved Common Cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: -0.25, 2.27\n",
      "|    ├- p-value: Not applicable\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Bootstrap Sample Dataset\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.253\n",
      "|    ├- p-value: 0.453\n",
      "|    └- Test significance: 2.23\n",
      "└-Refute: Use a Dummy Outcome\n",
      "     ├- Estimated effect: 0.00\n",
      "     ├- New effect: -0.008\n",
      "     ├- p-value: 0.436\n",
      "     └- Test significance: 0.00\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the model returns a summary of the identified causal effect and the refutation results. The `placebo_treatment_refuter` failed in this case because our intervention variable is binary. By default, all 6 potential refutation methods are run, but the user can also specify to only use a subset of them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (1/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (2/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (3/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (4/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (5/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (6/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (7/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (8/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (9/10)\n",
      "❗ Refutation 'placebo_treatment_refuter' returned invalid pval 'nan', retrying (10/10)\n",
      "Causal inference results for treatment variable 'aline_flg' and outcome variable 'icu_los_day':\n",
      "└- Increasing the treatment variable(s) [aline_flg] from 0 to 1 causes an increase of 2.2348813091683186 in the expected value of the outcome [['icu_los_day']], over the data distribution/population represented by the dataset.\n",
      "\n",
      "Refutation results\n",
      "├-Refute: Use a Placebo Treatment\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 0.000\n",
      "|    ├- p-value: nan\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Add a random common cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.234\n",
      "|    ├- p-value: 0.441\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Use a subset of data\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.243\n",
      "|    ├- p-value: 0.447\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Add an Unobserved Common Cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: -0.6, 2.25\n",
      "|    ├- p-value: Not applicable\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Bootstrap Sample Dataset\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.219\n",
      "|    ├- p-value: 0.454\n",
      "|    └- Test significance: 2.23\n",
      "└-Refute: Use a Dummy Outcome\n",
      "     ├- Estimated effect: 0.00\n",
      "     ├- New effect: 0.007\n",
      "     ├- p-value: 0.437\n",
      "     └- Test significance: 0.00\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        \"placebo_treatment_refuter\",  \n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"bootstrap_refuter\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By default, we are hiding a lot of the default `dowhy` output for clarity. However, it is possible to nontheless display it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "     d                                                                        \n",
      "────────────(E[icu_los_day|stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,c\n",
      "d[aline_flg]                                                                  \n",
      "\n",
      "                                   \n",
      "ad_flg,chf_flg,renal_flg,afib_flg])\n",
      "                                   \n",
      "Estimand assumption 1, Unconfoundedness: If U→{aline_flg} and U→icu_los_day then P(icu_los_day|aline_flg,stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,cad_flg,chf_flg,renal_flg,afib_flg,U) = P(icu_los_day|aline_flg,stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,cad_flg,chf_flg,renal_flg,afib_flg)\n",
      "\n",
      "## Realized estimand\n",
      "b: icu_los_day~aline_flg+stroke_flg+copd_flg+sepsis_flg+liver_flg+resp_flg+cad_flg+chf_flg+renal_flg+afib_flg\n",
      "Target units: ate\n",
      "\n",
      "## Estimate\n",
      "Mean value: 2.2348813091683186\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        # \"placebo_treatment_refuter\",  # we know it'll fail\n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"bootstrap_refuter\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    "    print_causal_estimate=True,\n",
    "    print_summary=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting \n",
    "Furthermore, the `causal_inference()` function can generate several plots, if desired. It's for example possible to output the causal graph ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        # \"placebo_treatment_refuter\",  # we know it'll fail\n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    "    print_causal_estimate=False,\n",
    "    print_summary=False,\n",
    "    show_graph=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "... or the graph generated by the add_unobserved_common_cause refutation method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "tags": [
     "hide-output"
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        # \"placebo_treatment_refuter\",  # we know it'll fail\n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    "    print_causal_estimate=False,\n",
    "    print_summary=False,\n",
    "    show_graph=False,\n",
    "    show_refute_plots=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the process of performing the causal `dowhy` analysis, an estimator is created which we can query from the model and then visualise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimate = ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        # \"placebo_treatment_refuter\",  # we know it'll fail\n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"bootstrap_refuter\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    "    print_causal_estimate=False,\n",
    "    print_summary=False,\n",
    "    show_graph=False,\n",
    "    show_refute_plots=False,\n",
    "    return_as=\"estimate\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'DoWhy estimate $\\\\rho$ (slope) = 2.235'}, xlabel='aline_flg', ylabel='icu_los_day'>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ep.pl.causal_effect(estimate)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Advanced options\n",
    "Within `dowhy`, the user can specify several arguments for identification, estimation and refutation. These arguments can also be passed directly to the respective functions through the `identify_kwargs`, `estimate_kwargs` and `refute_kwarg` of `causal_inference()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** Causal Estimate ***\n",
      "\n",
      "## Identified estimand\n",
      "Estimand type: EstimandType.NONPARAMETRIC_ATE\n",
      "\n",
      "### Estimand : 1\n",
      "Estimand name: backdoor\n",
      "Estimand expression:\n",
      "     d                                                                        \n",
      "────────────(E[icu_los_day|stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,c\n",
      "d[aline_flg]                                                                  \n",
      "\n",
      "                                   \n",
      "ad_flg,chf_flg,renal_flg,afib_flg])\n",
      "                                   \n",
      "Estimand assumption 1, Unconfoundedness: If U→{aline_flg} and U→icu_los_day then P(icu_los_day|aline_flg,stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,cad_flg,chf_flg,renal_flg,afib_flg,U) = P(icu_los_day|aline_flg,stroke_flg,copd_flg,sepsis_flg,liver_flg,resp_flg,cad_flg,chf_flg,renal_flg,afib_flg)\n",
      "\n",
      "## Realized estimand\n",
      "b: icu_los_day~aline_flg+stroke_flg+copd_flg+sepsis_flg+liver_flg+resp_flg+cad_flg+chf_flg+renal_flg+afib_flg\n",
      "Target units: days\n",
      "\n",
      "## Estimate\n",
      "Mean value: 2.2348813091683186\n",
      "\n",
      "Causal inference results for treatment variable 'aline_flg' and outcome variable 'icu_los_day':\n",
      "└- Increasing the treatment variable(s) [aline_flg] from 0 to 1 causes an increase of 2.2348813091683186 in the expected value of the outcome [['icu_los_day']], over the data distribution/population represented by the dataset.\n",
      "\n",
      "Refutation results\n",
      "├-Refute: Add a random common cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.234\n",
      "|    ├- p-value: 0.399\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Use a subset of data\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.228\n",
      "|    ├- p-value: 0.464\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Add an Unobserved Common Cause\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: -0.47, 2.24\n",
      "|    ├- p-value: Not applicable\n",
      "|    └- Test significance: 2.23\n",
      "├-Refute: Bootstrap Sample Dataset\n",
      "|    ├- Estimated effect: 2.23\n",
      "|    ├- New effect: 2.240\n",
      "|    ├- p-value: 0.482\n",
      "|    └- Test significance: 2.23\n",
      "└-Refute: Use a Dummy Outcome\n",
      "     ├- Estimated effect: 0.00\n",
      "     ├- New effect: 0.008\n",
      "     ├- p-value: 0.441\n",
      "     └- Test significance: 0.00\n",
      "\n"
     ]
    }
   ],
   "source": [
    "estimate = ep.tl.causal_inference(\n",
    "    adata=adata,\n",
    "    graph=causal_graph,\n",
    "    treatment=\"aline_flg\",\n",
    "    outcome=\"icu_los_day\",\n",
    "    estimation_method=\"backdoor.linear_regression\",\n",
    "    refute_methods=[\n",
    "        # \"placebo_treatment_refuter\",  # we know it'll fail\n",
    "        \"random_common_cause\",\n",
    "        \"data_subset_refuter\",\n",
    "        \"add_unobserved_common_cause\",\n",
    "        \"bootstrap_refuter\",\n",
    "        \"dummy_outcome_refuter\",\n",
    "    ],\n",
    "    print_causal_estimate=True,\n",
    "    print_summary=True,\n",
    "    show_graph=False,\n",
    "    show_refute_plots=\"contour\",\n",
    "    return_as=\"estimate\",\n",
    "    identify_kwargs={\"proceed_when_unidentifiable\": True},\n",
    "    estimate_kwargs={\"target_units\": \"days\"},\n",
    "    refute_kwargs={\"random_seed\": 5},\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.4"
  },
  "vscode": {
   "interpreter": {
    "hash": "7f986eb25c8d585612eeb3751d1372c254a3aec44495756539506c5b5c1a3ad8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}