{ "abstract": "Background: Using the 2025 Chikungunya fever (CHIK) outbreak in Chancheng District, Foshan City, this study applied a random forest (RF) regression model combined with SHapley Additive exPlanations (SHAP) to explore predictive associations, nonlinear relationships and potential thresholds between environmental–social factors and village/community-level cumulative incidence, to inform stratified control of mosquito-borne diseases. Methodology/principal findings: In this cross-sectional ecological study of 143 villages/communities, the outcome was cumulative incidence, and nine candidate covariates were assessed, including the hospitalization isolation rate, construction-site density, and population density. Multicollinearity was checked using the variance inflation factor. Model fit was evaluated by the out-of-bag (OOB) R², RMSE, and MAE, and variable importance by %IncMSE. Robustness was tested with 100 repeated runs, bootstrap thresholds from SHAP dependence plots, and a sensitivity analysis excluding the endogenous isolation rate. On the log(1 + incidence) scale, the OOB R² was 0.206, with underestimation of high-incidence areas. The hospitalization isolation rate had the highest importance (%IncMSE = 18.43) and was negatively correlated with predictions (ρ = −0.749), but this likely reflects reverse causation and is predictive only. Construction-site density was strongly positive (ρ = 0.855), with a stable threshold near 12.8 sites/km²; it remained the most robust predictor after removing the isolation rate (%IncMSE = 8.54). Conclusions/significance: Construction-site density was the most robust environmental predictor, whereas population density contributed little. These exploratory, predictive associations—not causal effects—should guide risk stratification and require prospective validation with time-matched longitudinal data. Author summary: Chikungunya fever is a mosquito-borne illness causing fever and prolonged joint pain. In recent years, outbreaks have grown in scale worldwide, posing serious public health challenges. In 2025, Foshan City in southern China experienced the largest locally acquired chikungunya outbreak ever recorded in the country. We focused on Chancheng District, the urban core of Foshan, where 1404 cases were reported across 143 villages and communities. Our question was: why did some communities have far more cases than others, even within the same district. Using an interpretable machine-learning approach (random forest with SHAP analysis), we examined which environmental and social factors predicted community-level incidence. The density of small construction sites—which readily accumulate discarded water-holding containers where mosquitoes breed—was the most stable environmental predictor: risk climbed sharply only after crossing a clear threshold (about 13 sites per square kilometer), rather than rising gradually. By contrast, overall population density had little predictive value in this already crowded setting, suggesting that fine-scale environmental exposure matters more than population size. The hospitalization isolation rate showed the strongest statistical association with lower incidence, but we interpret it cautiously: communities that brought their outbreaks under control tended to reach higher isolation rates, so this likely reflects the response to the epidemic rather than proof that isolation reduced transmission. As an exploratory analysis based on data from a single time point, our study points to predictive associations useful for flagging high-risk communities and targeting limited control resources—not causal effects—and prospective studies are needed to confirm them.", "abstract_shown": true, "affiliations": [ "Chancheng District Center for Disease Control and Prevention, Foshan, Guangdong, China", "Chinese Field Epidemiology Training Program, Chinese Center for Disease Control and Prevention (Chinese Academy of Preventive Medicine), Beijing, China", "Shunyi District Center for Disease Control and Prevention, Beijing, China", "The Johns Hopkins University, Baltimore, Maryland, United States of America" ], "article_type": "Research Article", "arxiv": null, "author_total": 11, "authors": [ { "affs": [ 0 ], "name": "Fengling Chen" }, { "affs": [ 1 ], "name": "Jinsen He" }, { "affs": [ 2 ], "name": "Huihui Liu" }, { "affs": [ 3 ], "name": "Zhilin Wu" }, { "name": "Jiwen Wu" }, { "name": "Kunlu He" }, { "name": "Hongjuan Wen" }, { "name": "Xiaohua Li" }, { "name": "Hongwei Wu" }, { "name": "Lang Lang" }, { "name": "Lijie Zhang" } ], "bibtex": "https://research.draconitic.com/paper/dk65jwc4vyxpp.bib", "cited_by": 0, "doi": "10.1371/journal.pntd.0014675", "fulltext_url": "https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0014675", "id": "dk65jwc4vyxpp", "issns": [ "1935-2735" ], "landing_url": "https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0014675", "language": null, "licence": "CC-BY-4.0", "licence_label": "CC BY 4.0", "licence_source": "journal", "licence_url": "https://creativecommons.org/licenses/by/4.0/", "metadata_licence": null, "native_subjects": [ [ "openalex", "2739|Mosquito-borne diseases and control" ], [ "plos", "/Biology and life sciences/Organisms/Bacteria/Wolbachia" ], [ "plos", "/Biology and life sciences/Organisms/Eukaryota/Animals/Invertebrates/Arthropoda/Insects/Mosquitoes" ], [ "plos", "/Biology and life sciences/Population biology/Population metrics/Population density" ], [ "plos", "/Biology and life sciences/Species interactions/Disease vectors/Insect vectors/Mosquitoes" ], [ "plos", "/Biology and life sciences/Zoology/Animals/Invertebrates/Arthropoda/Insects/Mosquitoes" ], [ "plos", "/Biology and life sciences/Zoology/Entomology/Insects/Mosquitoes" ], [ "plos", "/Medicine and health sciences/Epidemiology" ], [ "plos", "/Medicine and health sciences/Epidemiology/Medical risk factors" ], [ "plos", "/Medicine and health sciences/Health care/Health care facilities/Hospitals/Hospitalizations" ], [ "plos", "/Medicine and health sciences/Medical conditions/Infectious diseases/Disease vectors/Insect vectors/Mosquitoes" ], [ "plos", "/Medicine and health sciences/Medical conditions/Infectious diseases/Infectious disease control" ], [ "plos", "/Medicine and health sciences/Medical conditions/Infectious diseases/Viral diseases/Chikungunya infection" ], [ "plos", "/Medicine and health sciences/Medical conditions/Tropical diseases/Neglected tropical diseases/Chikungunya infection" ] ], "openalex": "W7207569827", "pdf_url": "https://journals.plos.org/plosntds/article/file?id=10.1371/journal.pntd.0014675&type=printable", "pmcid": null, "pmid": null, "published": "2026-09-03", "publisher": "Public Library of Science", "ris": "https://research.draconitic.com/paper/dk65jwc4vyxpp.ris", "sources": [ { "first_seen": "2026-10-30T12:17:56Z", "native_id": "10.1371/journal.pntd.0014675", "source": "plos", "url": "https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0014675" } ], "subjects": [ { "field": "biology", "origin": "plos:zoology", "subfield": "zoology-behaviour" }, { "field": "medicine", "origin": "plos:epidemiology", "subfield": "epidemiology-public-health" }, { "field": "medicine", "origin": "plos:infectious diseases", "subfield": "infectious-disease" } ], "title": "Predictive association between control measures and chikungunya fever incidence based on random forest and SHAP analysis", "title_text": "Predictive association between control measures and chikungunya fever incidence based on random forest and SHAP analysis", "type": "peer-reviewed", "updated_at": "2026-10-30T12:17:56Z", "url": "https://research.draconitic.com/paper/dk65jwc4vyxpp", "venue": "PLOS Neglected Tropical Diseases", "versions": [] }