Analyzing Citizens’ and Health Care Professionals’ Searches for Smell/Taste Disorders and Coronavirus in Finland During the COVID-19 Pandemic: Infodemiological Approach Using Database Logs
Bibliographic record
Abstract
BACKGROUND: The COVID-19 pandemic has prevailed over a year, and log and register data on coronavirus have been utilized to establish models for detecting the pandemic. However, many sources contain unreliable health information on COVID-19 and its symptoms, and platforms cannot characterize the users performing searches. Prior studies have assessed symptom searches from general search engines (Google/Google Trends). Little is known about how modeling log data on smell/taste disorders and coronavirus from the dedicated internet databases used by citizens and health care professionals (HCPs) could enhance disease surveillance. Our material and method provide a novel approach to analyze web-based information seeking to detect infectious disease outbreaks. OBJECTIVE: The aim of this study was (1) to assess whether citizens' and professionals' searches for smell/taste disorders and coronavirus relate to epidemiological data on COVID-19 cases, and (2) to test our negative binomial regression modeling (ie, whether the inclusion of the case count could improve the model). METHODS: We collected weekly log data on searches related to COVID-19 (smell/taste disorders, coronavirus) between December 30, 2019, and November 30, 2020 (49 weeks). Two major medical internet databases in Finland were used: Health Library (HL), a free portal aimed at citizens, and Physician's Database (PD), a database widely used among HCPs. Log data from databases were combined with register data on the numbers of COVID-19 cases reported in the Finnish National Infectious Diseases Register. We used negative binomial regression modeling to assess whether the case numbers could explain some of the dynamics of searches when plotting database logs. RESULTS: We found that coronavirus searches drastically increased in HL (0 to 744,113) and PD (4 to 5375) prior to the first wave of COVID-19 cases between December 2019 and March 2020. Searches for smell disorders in HL doubled from the end of December 2019 to the end of March 2020 (2148 to 4195), and searches for taste disorders in HL increased from mid-May to the end of November (0 to 1980). Case numbers were significantly associated with smell disorders (P<.001) and taste disorders (P<.001) in HL, and with coronavirus searches (P<.001) in PD. We could not identify any other associations between case numbers and searches in either database. CONCLUSIONS: Novel infodemiological approaches could be used in analyzing database logs. Modeling log data from web-based sources was seen to improve the model only occasionally. However, search behaviors among citizens and professionals could be used as a supplementary source of information for infectious disease surveillance. Further research is needed to apply statistical models to log data of the dedicated medical databases.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.012 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.007 | 0.005 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".