Public Search Behavior and Tuberculosis Cases in Indonesia 2019-2023: An Infodemiology Method Using Google Trends

This dataset contains the raw data used in the study titled:
“Public Search Behavior and Tuberculosis Cases in Indonesia 2019–2023: An Infodemiology Method Using Google Trends.”
The dataset provides province-level data (34 provinces in Indonesia) covering the period 2019–2023, integrating:
1. Tuberculosis Case DataAnnual number of reported TB cases per province (2019–2023)TB case scaling variables (standardized values used for analysis)2. Google Trends Search DataRelative Search Volume (RSV; scale 0–100) for multiple tuberculosis-related keywords, grouped into thematic categories:
General terms: TB, TBC, Tuberkulosis, TB ParuSymptoms-related terms: Gejala TB, Gejala TBC, etc.Signs/characteristics: Ciri TB, Ciri TBC, etc.Treatment-related terms: Obat TB, Obat TBC, etc.Prevention-related terms: Pencegahan TB, etc.Transmission-related terms: Penularan TB, etc.Layman/local terms: Flek paru, Flek paru-paruPopulation-specific terms: TB anak, TB bayi, and related variationsEach keyword is recorded annually at the provincial level using Google Trends normalized indices.
3. Data StructureUnit of analysis: Province-yearSpatial coverage: Indonesia (all provinces)Temporal coverage: 2019–2023Format: Wide-format dataset with yearly blocks of variables4. Data SourcesTuberculosis case data: Indonesia Health Profile Reports (Profil Kesehatan Indonesia), Ministry of HealthSearch data: Google Trends (https://trends.google.com), retrieved using standardized keyword queries5. Notes for UsersGoogle Trends values are relative (0–100) and do not represent absolute search countsDifferences across years reflect relative popularity within each query period, not direct comparability without normalizationThe dataset is intended for infodemiology, correlation, and ecological analysis6. ReproducibilityThis dataset enables full replication of:
Correlation analysis between TB cases and search behaviorTrend comparison across provinces and yearsKeyword-level infodemiology modeling