Abstract / Summary
This deposit contains the Google Trends export files exactly as downloaded, the download verification record, the cleaned analysis dataset, and all analysis code for the study named above. It allows an independent reader to reproduce every number, table and figure in the article from the original downloads. Study design. Retrospective infodemiological study of Google Trends relative search volume (RSV) for five facial aesthetic search entities across nine countries and a worldwide aggregate, from January 2004 to August 2026. Each series holds 272 consecutive monthly observations, giving 13,600 observations in total. Rhinoplasty, blepharoplasty, rhytidectomy and injectable filler were queried as Google Trends Topics; Botox was queried as a literal search term. All five were queried together in one panel per area so that they share a single internal normalisation. Contents.- raw_google_trends_exports/final_panel/: the 10 files used in the analysis, byte for byte as downloaded on 5 September 2026.- raw_google_trends_exports/pilot_and_excluded/: 6 pilot files that are not analysed. They document two protocol changes the article discloses: chin augmentation and the botulinum toxin Topic were replaced by injectable filler and the Botox search term, and Nigeria was replaced by South Africa under a prespecified exclusion rule.- raw_google_trends_exports/download_verification.csv: per-file record made at download time, including SHA-256, extraction timestamp, row count and structural checks.- data/: the cleaned analysis dataset in long and wide format, plus the cleaning notes.- Seven Python scripts covering the main analysis, the seasonal and correlation analyses, figure generation, article-format tables, and an independent verification script.- CHECKSUMS.sha256: SHA-256 for every deposited file.- README.md: file-by-file descriptions, exact software versions, and the order in which to run the scripts. Provenance. The SHA-256 of all 10 analysed files was recomputed when this deposit was assembled and matches the value recorded at download time for every file. No relative search volume was altered, imputed or rescaled during cleaning, and zeros were kept as the platform reported them. Software. Python 3.9.6 with pandas 2.3.3, numpy 2.0.2, scipy 1.13.1, statsmodels 0.14.6, scikit-posthocs 0.16.1, matplotlib 3.9.4 and openpyxl 3.1.5. Exact versions are pinned in requirements.txt. The scripts resolve all paths relative to their own location, so the deposit runs from any directory. The one stochastic step, a moving-block bootstrap, is seeded and reproduces exactly. Reproduction. The deposit was unpacked into an empty directory and run end to end before release. All seven scripts completed and the regenerated tables are identical to those behind the article. Note on relative search volume. Google Trends does not report counts. It scales searches against total search activity in the selected place and period, then sets 100 at the peak of that scaled series for the leading term in the panel. Values are comparable within an area and not comparable in absolute magnitude between areas. A zero means search volume fell below the reporting threshold. Google Trends is a live service and a later download can return slightly different historical values, which is why dated files and checksums are deposited rather than a download instruction.