{ "cells": [ { "cell_type": "code", "execution_count": 4, "id": "86a20e4d", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "from statsmodels.stats.weightstats import DescrStatsW\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 5, "id": "d396ac72", "metadata": {}, "outputs": [], "source": [ "# Weighted mean & SD\n", "def w_mean_sd(series, weights):\n", " s = DescrStatsW(series, weights=weights, ddof=0)\n", " return s.mean, s.std\n", "\n", "# Weighted % for categorical\n", "def w_percent(df, var, target=None):\n", " d = df.copy()\n", " if target is not None:\n", " d = d[d['IRAMDEYR']==target]\n", " return (d.groupby(var)['ANALWT2_C'].sum() / d['ANALWT2_C'].sum() * 100).round(1)\n", "\n", "# Weighted N (population estimate)\n", "def weighted_n(df):\n", " return round(df['ANALWT2_C'].sum() / 1e6, 1) # in million (approx population)" ] }, { "cell_type": "code", "execution_count": 6, "id": "16fa1242", "metadata": {}, "outputs": [], "source": [ "# Columns you need\n", "cols = [\n", " 'IRAMDEYR','CATAG7','IRSEX','NEWRACE2',\n", " 'AKSSLR6WRST','WHODASDASC','UD5ILLANY','ANALWT2_C'\n", "]\n", "\n", "# Load dataset (tab-delimited)\n", "df = pd.read_csv(\"../data/raw/NSDUH_2023_Tab.txt\", sep='\\t', usecols=cols, low_memory=False)\n", "\n", "# Adults only (CATAG7 ≥ 4)\n", "df = df[df['CATAG7'] >= 4]\n", "\n", "# Drop missing target\n", "df = df[df['IRAMDEYR'].notna()]\n" ] }, { "cell_type": "code", "execution_count": 7, "id": "95d99ad9", "metadata": {}, "outputs": [], "source": [ "no_mde_raw = len(df[df.IRAMDEYR == 0])\n", "mde_raw = len(df[df.IRAMDEYR == 1])\n", "total_raw = no_mde_raw + mde_raw\n", "\n", "# Percent share of total (based on sample)\n", "no_mde_pct = round(no_mde_raw / total_raw * 100, 1)\n", "mde_pct = round(mde_raw / total_raw * 100, 1)\n", "\n" ] }, { "cell_type": "code", "execution_count": 8, "id": "096da4be", "metadata": {}, "outputs": [], "source": [ "# ---------- Build rows ----------\n", "rows = []\n", "\n", "# --- 1️⃣ Total participants row ---\n", "rows.append([\n", " \"Total participants\",\n", " \"Total respondents in the analytic sample\",\n", " f\"{no_mde_raw:,} ({no_mde_pct:,} %)\",\n", " f\"{mde_raw:,} ({mde_pct:,} %)\"\n", "])\n", "\n", "# --- 2️⃣ Continuous variables (mean ± SD) ---\n", "for var in ['AKSSLR6WRST','WHODASDASC']:\n", " mean0, sd0 = w_mean_sd(df[df.IRAMDEYR==0][var], df[df.IRAMDEYR==0]['ANALWT2_C'])\n", " mean1, sd1 = w_mean_sd(df[df.IRAMDEYR==1][var], df[df.IRAMDEYR==1]['ANALWT2_C'])\n", " rows.append([var, \"\", f\"{mean0:.2f} ± {sd0:.2f}\", f\"{mean1:.2f} ± {sd1:.2f}\"])\n", "\n", "# --- 3️⃣ Categorical variables (weighted %) ---\n", "for var in ['CATAG7','IRSEX','NEWRACE2','UD5ILLANY']:\n", " no_mde = w_percent(df, var, 0)\n", " mde = w_percent(df, var, 1)\n", " for level in sorted(no_mde.index):\n", " rows.append([\n", " f\"{var}_{level}\",\n", " \"\",\n", " f\"{no_mde.get(level, np.nan)} %\",\n", " f\"{mde.get(level, np.nan)} %\"\n", " ])" ] }, { "cell_type": "code", "execution_count": 9, "id": "ed2bf402", "metadata": {}, "outputs": [], "source": [ "# ---------- Convert to DataFrame ----------\n", "table1 = pd.DataFrame(rows, columns=[\n", " \"Feature (code)\",\n", " \"Code Meaning\",\n", " \"No MDE (Weighted %) or Mean ± SD\",\n", " \"MDE (Weighted %) or Mean ± SD\"\n", "])\n" ] }, { "cell_type": "code", "execution_count": 10, "id": "27184f74", "metadata": {}, "outputs": [], "source": [ "# ---------- Add readable code meanings ----------\n", "labels = {\n", " 'CATAG7_4':'Age 18–25 y','CATAG7_5':'Age 26–34 y',\n", " 'CATAG7_6':'Age 35–49 y','CATAG7_7':'Age 50+ y',\n", " 'IRSEX_1':'Male','IRSEX_2':'Female',\n", " 'NEWRACE2_1':'White','NEWRACE2_2':'Black or African American',\n", " 'NEWRACE2_3':'American Indian / Alaska Native',\n", " 'NEWRACE2_4':'Native Hawaiian / Pacific Islander',\n", " 'NEWRACE2_5':'Asian','NEWRACE2_6':'Two or more races','NEWRACE2_7':'Hispanic or Latino',\n", " 'UD5ILLANY_0':'No Drug Use Disorder','UD5ILLANY_1':'Has Drug Use Disorder',\n", " 'AKSSLR6WRST':'K6 Psychological Distress Score (0–24)',\n", " 'WHODASDASC':'WHODAS Disability Score (0–100)'\n", "}\n", "\n", "table1[\"Code Meaning\"] = table1[\"Feature (code)\"].map(labels).fillna(table1[\"Code Meaning\"])\n" ] }, { "cell_type": "code", "execution_count": 11, "id": "e8c06b9b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "✅ Clean Table 1 saved as 'table1_Baseline_Characteristics.csv'\n", " Feature (code) Code Meaning \\\n", "0 Total participants Total respondents in the analytic sample \n", "1 AKSSLR6WRST K6 Psychological Distress Score (0–24) \n", "2 WHODASDASC WHODAS Disability Score (0–100) \n", "3 CATAG7_4 Age 18–25 y \n", "4 CATAG7_5 Age 26–34 y \n", "5 CATAG7_6 Age 35–49 y \n", "6 CATAG7_7 Age 50+ y \n", "7 IRSEX_1 Male \n", "8 IRSEX_2 Female \n", "9 NEWRACE2_1 White \n", "\n", " No MDE (Weighted %) or Mean ± SD MDE (Weighted %) or Mean ± SD \n", "0 39,948 (88.5 %) 5,185 (11.5 %) \n", "1 1.19 ± 2.96 9.01 ± 5.20 \n", "2 0.79 ± 1.78 4.93 ± 2.66 \n", "3 4.3 % 8.4 % \n", "4 7.6 % 18.3 % \n", "5 14.8 % 24.4 % \n", "6 73.2 % 48.9 % \n", "7 49.9 % 35.4 % \n", "8 50.1 % 64.6 % \n", "9 60.7 % 66.0 % \n" ] } ], "source": [ "# ---------- Export clean final table ----------\n", "table1.to_csv(\"../results/table1_Baseline_Characteristics.csv\", index=False)\n", "print(\"✅ Clean Table 1 saved as 'table1_Baseline_Characteristics.csv'\")\n", "print(table1.head(10))" ] } ], "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": "3.8.10" } }, "nbformat": 4, "nbformat_minor": 5 }