vaccine_scalars

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scalar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ S = SEV*(RR(max) - 1) + 1 $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### risk-acause specific PAF:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ PAF = 1 - \\frac{1}{S} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SEV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Since vaccination is a dichotomous risk factor, so the SEV of non-vaccinated is calculated as 1 - vaccine coverage."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RR(max)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Since vaccination is a dichotomous risk factor, so the RR(max) of non-vaccinated is equivalent to RR. \n",
    "* Simulate 1000 draws of coefficients based on normal distribution using coefficients provided. \n",
    "* Convert draws of coefficients to draws of RR based on the formula below depending on modeler's methods and results for each cause."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (diptheria, dtp3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of dtp3: /ihme/forecasting/data/vaccine_coverage/coverage/best/dtp3_draws.h5 (from Mollie)\n",
    "* coefficients: one set of coefficient and standard error\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/diptheria_vaccine_coef.csv (from Hmwe (Diphtheria negative binomial model), formatted by Jiawei)\n",
    "    * DTP3_coverage_prop = proportion vaccinated with DTP3\n",
    "    * age_group_id = GBD 2015 age group ids"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ RR = \\frac{1}{e^{coeff}} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (measles, measles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of measles: /ihme/forecasting/data/vaccine_coverage/coverage/best/measles_draws.h5 (from Mollie)\n",
    "* coefficients: one set of coefficient and standard error\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/measles_vaccine_coef.csv (from Hmwe (Measles incidence model), formatted by Jiawei)\n",
    "    * ln_unvacc = proportion unvaccinated with measles vaccine (log transformed)\n",
    "    * supp_lagX = supplementary vaccination coverage lagged by X year"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ R = e^{e^{coeff}} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (whooping, dtp3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of dtp3: /ihme/forecasting/data/vaccine_coverage/coverage/best/dtp3_draws.h5 (from Mollie)\n",
    "* coefficients: one set of coefficient and standard error\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/whooping_vaccine_coef.csv (from Hmwe (Pertussis incidence model), formatted by Jiawei)\n",
    "    * ln_unvacc= proportion unvaccinated with DTP3 (log transformed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ R = e^{e^{coeff}} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (lri_hib, dtp3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "* exposure (1 - sev) of dtp3: /ihme/forecasting/data/vaccine_coverage/coverage/best/dtp3_draws.h5 (from Mollie)\n",
    "* coefficients: one set of (effect size, effect lower, effect upper)\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/hib_effect_size.xlsx (from Chris Troeger, formatted by Jiawei)\n",
    "    *  The Hib file provides the vaccine effectiveness of the Hib vaccine against all pneumonia. These numbers are the same by age, geography, and time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ RR = \\frac{1}{1-effectsize} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (lri_pneumo, PCV)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of PCV: /ihme/forecasting/data/vaccine_coverage/coverage/best/pcv_draws.h5 (from Mollie)\n",
    "* coefficients: one set of (effect size, lower, upper) for PCV7/PCV10/PCV13\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/pcv_effect.xlsx (from Chris Troeger)\n",
    "* PCV type info: /ihme/forecasting/data/vaccine_coverage/scalars/inputs/pcv_introduction_supplementary_info.xlsx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ RR = \\frac{1}{1-effectsize} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* There are three types of PCV vaccine(7/10/13), most countries have at least one type vaccinated. For countries that \n",
    "    do not have information, use coefficients of PCV10."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## (diarrhea_rotavirus, rota)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of rota: /ihme/forecasting/data/vaccine_coverage/coverage/best/rota_draws.h5 (from Mollie)\n",
    "* coefficients: age-specific odds ratios and standard errors\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/rotavirus_odds_ratios.xlsx (from Chris Troeger, formatted by Jiawei)\n",
    "    * The rotavirus file provides the odds ratio of diarrhea given rotavirus in a stool sample, by age. These numbers are the same by geography and time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "odds ratio is relative risk"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (tetanus, dtp3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* exposure (1 - sev) of dtp3: /ihme/forecasting/data/vaccine_coverage/coverage/best/dtp3_draws.h5 (from Mollie)\n",
    "* coefficients: age-specific, country-specific, sex-specific coefficients and standard errors\n",
    "    * /ihme/forecasting/data/vaccine_coverage/scalars/inputs/submodel_coeffs_err_only_log.csv\n",
    "    * Based on model_version_id, different demographies have different coefficents.\n",
    "    * Demography information can be obtained from database cod-dbsnapshot-d01, table model_version"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### formula"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$ R = e^{{coeff}} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### outputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* pafs: /ihme/forecasting/data/paf/best/risk_acause\\_specific/{acause}\\_{risk}_{sex_id}.hdf"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}