Mpox on Instagram: A content analytic study (Preprint)
Bibliographic record
Abstract
Abstract Background Mpox was declared a public health emergency of international concern in 2022. Instagram is widely used by age groups and communities disproportionately affected by mpox; yet platform-specific evidence on mpox information characteristics and engagement remains limited. Objective The aim of this study is to characterize sources, content, and engagement features of mpox-related Instagram posts, to describe prevention and treatment framing, and to compare the top 10% most-liked posts with the remaining corpus. Methods We retrieved English-language public Instagram posts via CrowdTangle containing “mpox” or “monkeypox” dated May 5, 2022, to January 17, 2023 (initial N=18,616). Using a pretested, deductive codebook adapted from prior Instagram health studies, 2 coders completed 2 pilot rounds; variables with low agreement were excluded. A randomized analytic sample of 1000 posts was coded for source type, content features, and prevention/treatment framing. Descriptive statistics were computed. For engagement contrasts, we compared the top 10% most-liked posts with the bottom 90% using tests of differences in independent proportions (mean differences [MD] with P values). Results Most posts originated from organizations (760/1000, 76%) versus individuals (240/1000, 24%). Organizational sources most commonly included businesses (436/760, 57.4%) and news/media outlets (401/760, 52.8%); government (174/760, 22.9%), nonprofits (131/760, 17.3%), and health care organizations (70/760, 9.2%) were less frequent. About one-third of posts cited a source (344/1000, 34.4%), most often the World Health Organization (WHO) and Centers for Disease Control and Prevention (CDC)/other federal entity. Posts predominantly used illustrated images/infographics (827/1000, 82.7%); photos appeared in 47.3% (473/1000) and videos in 12.4% (124/1000) of posts. Prevention content appeared in 38.4% (384/1000) of posts, most commonly vaccination (684/1000, 68.5% of prevention posts), followed by avoiding close contact (145/1000, 14.5%), avoiding contact with objects (83/1000, 8.3%), abstaining from sexual activity (76/1000, 7.6%), and condom use (13/1000, 1.3%); 28.9% (289/1000) of prevention posts noted barriers. Treatment mentions were uncommon (25/1000, 2.5% traditional biomedical; 2/1000, 0.2% alternative). Compared with the bottom 90%, the top 10% most-liked posts (1) were more likely to originate from public figures/celebrities among individuals (MD=−0.591; P <.001) and from businesses (MD=−0.299; P <.001) or news/media (MD=−0.350; P <.001) among organizations; (2) were less likely to be from government ( P <.001) , nonprofit ( P =.006), or health care organizations ( P =.005); and (3) more often included nonmoving images (MD=−0.119; P =.024), visible lesion depictions (MD=−0.081; P =.035), prevalence mentions (MD=−0.180; P <.001), and citations (MD=−0.162; P =.001). Conclusions During the initial outbreak period, the highly engaged mpox content on Instagram skewed toward posts by public figures and news/business accounts and toward static, citation-bearing visuals that included prevalence context and occasionally lesion imagery. Public-health communicators seeking reach on Instagram should prioritize clear static infographics with explicit source citation and epidemiologic context and consider copublishing with trusted creators and news outlets, while addressing access barriers highlighted in prevention posts.
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 | 0.005 | 0.028 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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".