Where Malaria Could Return: Scientists Map the Riskiest Districts in a Malaria-Free Chinese City

China was certified malaria-free by the World Health Organization in 2021, a milestone built on decades of surveillance, drug distribution and mosquito control. Yet elimination is not the same as eradication, and the parasite has not left the planet. Every year, travelers and overseas workers return to Chinese cities carrying Plasmodium in their blood, and across much of the country the Anopheles mosquitoes capable of transmitting that parasite still thrive. In the southwestern municipality of Chongqing, a team of public health researchers has now combined three independent lines of evidence—imported case records, habitat modeling and field-measured mosquito density—to identify precisely which districts face the greatest risk of malaria quietly taking hold again.

The study, led by Dan Jiang of the Chongqing Center for Disease Control and Prevention together with colleagues at the national institute of parasitic diseases in Shanghai, compiled every imported malaria case reported in Chongqing between 2014 and 2024. That amounted to 301 cases, with annual numbers fluctuating between 4 and 48. The predominant species was Plasmodium falciparum, the deadliest human malaria parasite, which accounted for 75.08 percent of infections, followed by P. vivax at 12.29 percent, P. ovale at 7.97 percent, and smaller numbers of P. malariae and mixed infections. Four patients developed severe malaria and three died, a reminder that even in a non-endemic setting, imported malaria remains a life-threatening condition.

The demographic profile of those infected tells a clear story about who bears the risk. Nearly 94.35 percent of cases were male, 84.71 percent were aged 21 to 50, and 60.80 percent were overseas laborers—workers, migrants and farmers employed abroad. The infections overwhelmingly traced back to Africa, with 90.37 percent of cases originating in 32 African countries; the Democratic Republic of the Congo and Nigeria each contributed 44 cases, and Guinea added 29. This pattern reflects Chongqing’s role as a key node city along China’s Belt and Road Initiative, which has expanded economic cooperation with malaria-endemic regions and sent a steady stream of young, working-age men to construction sites and enterprises in sub-Saharan Africa. Only 22 cases were recorded during the COVID-19 pandemic years of 2020 to 2022, when travel collapsed, but numbers rebounded sharply afterward, peaking at 48 in 2024—a signal that population mobility, not local ecology, drives importation.

Diagnosis often came late. The median interval from symptom onset to diagnosis was 5.6 days, and 41.20 percent of patients were only diagnosed more than a week after falling ill. The authors attribute this to low malaria awareness among returning workers and limited diagnostic capacity at county-level hospitals, where nearly half of the cases first sought care. Because falciparum malaria can progress from fever to organ failure within days, such delays carry real clinical danger, and the researchers recommend routine malaria screening for febrile patients at county hospitals, along with pre-departure education and post-arrival active screening for returning migrant workers.

But case counts alone cannot predict where transmission might restart. For malaria to re-establish locally, three conditions must coincide: imported parasites, competent vectors and environmental conditions that allow mosquitoes to survive and breed in sufficient numbers. Chongqing historically was one of China’s most malaria-burdened regions, with all of its districts and counties reporting cases—predominantly vivax malaria—before transmission was brought under control in the 1990s. No locally acquired case has been reported since 2011, and the municipality passed its national elimination assessment in 2019. The dominant vector today is Anopheles sinensis, a species found in every surveyed district between 2011 and 2025, whose larvae flourish in rice paddies, irrigation ditches, streams and marshes, and which has developed widespread resistance to pyrethroid insecticides, the backbone of standard vector control.

To map where this mosquito could persist, the team built an ensemble ecological niche model using the biomod2 framework, which integrates twelve different modeling algorithms—including random forests, gradient boosting, generalized linear models and maximum entropy—into a single consensus prediction. Occurrence records collected from 2011 to 2025 were spatially thinned so that no more than one point remained per square kilometer, yielding 127 valid records and reducing the overfitting risk from spatial autocorrelation. Environmental predictors were screened by Pearson correlation, and nine variables entered the final models, drawn from WorldClim bioclimatic data at 30-arc-second resolution and vegetation indices from NASA’s MODIS satellite product. In total, 562 individual models were fitted, each trained on 80 percent of the data and tested on the remainder, with only high-performing runs retained for the ensemble.

The ensemble model achieved an area under the receiver operating characteristic curve of 0.839 and a true skill statistic of 0.531, outperforming most individual algorithms. Among the environmental drivers, three variables dominated: bio5, the maximum temperature of the warmest month; NDVI, the normalized difference vegetation index, which captures vegetation greenness and soil moisture; and bio11, the mean temperature of the coldest quarter. These variables make biological sense. Moderate vegetation cover offers resting and egg-laying sites for adult mosquitoes, while mild winters in western Chongqing favor overwintering survival and early spring population resurgence. The resulting suitability map showed that most of Chongqing offers at least some suitable habitat, with highly suitable areas concentrated in the metropolitan core and western districts, and suitability dropping off in the high, cold mountains of the east, such as Chengkou and Wuxi.

Crucially, the researchers recognized that climatic suitability approximates only the mosquito’s fundamental niche, not its actual abundance, which depends on local factors like breeding-site availability and land use. So they overlaid field-measured vector density—captures per trap-night averaged over the June-to-October peak seasons from 2011 to 2024—onto the suitability map. The result was striking heterogeneity: the highest densities occurred in Wanzhou, followed by Yunyang, Fengjie and Fengdu, all major rice-growing areas whose extensive paddies sustain large mosquito populations, while the metropolitan districts showed uniformly low densities except in Jiulongpo. In six districts, including Beibei and Bishan, no An. sinensis were captured at all despite suitable climate, illustrating how urbanization and sustained control can suppress even a well-adapted vector.

The final step examined imported P. vivax cases specifically, because An. sinensis is the primary vector for that species in the region. Of the 39 imported vivax cases, concentrations appeared in Yubei, Hechuan and Pengshui—districts connected to international travel and labor migration, with Yubei hosting Jiangbei International Airport—and notably, 51.28 percent of these cases arrived during the peak mosquito activity season. When case importation, habitat suitability and measured vector density were all overlaid, six districts emerged as top priorities: Wanzhou, Yubei, Hechuan, Tongnan, Pengshui and Jiulongpo. Eleven more, including Tongliang, Shapingba, Jiangjin, Yongchuan, Dazu, Rongchang, Fuling, Wulong, Xiushan, Zhongxian and Wushan, were classified as moderate priority.

The framework is not without limitations, which the authors acknowledge candidly. Two historically important vectors, An. minimus and An. lesteri, were never detected in surveillance and were excluded, though they could persist at undetectable levels; the modest number of vivax cases limited quantitative risk estimation; and the district-level analysis may mask finer township-scale variation. Still, the study offers one of the first municipal-level integrations of case surveillance, niche modeling and empirical vector density in the post-elimination era, and its practical prescription is straightforward: reinforce microscopy capacity at primary clinics in high-priority districts, actively screen returning migrant workers, and concentrate seasonal vector suppression during the summer-autumn months when imported vivax cases and mosquito abundance are most likely to collide. For a city celebrating more than a decade without local transmission, the message is clear—vigilance, targeted by data, is the price of keeping malaria out for good.

Subject of Research: Integrated risk mapping of imported malaria re-establishment in Chongqing, China

Article Title: An integrated approach to prioritizing areas for imported malaria re-establishment prevention in Chongqing, China

Article References: Jiang, D., Tan, Y., Yang, M., Chen, S., Guan, H., Yang, S., Bai, H., Luo, F., Chen, J.-H., & Zhou, S. (2026). An integrated approach to prioritizing areas for imported malaria re-establishment prevention in Chongqing, China. Frontiers in Cellular and Infection Microbiology, 16, Article 1929309. https://doi.org/10.3389/fcimb.2026.1929309

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1929309

Keywords: imported malaria, malaria re-establishment, Anopheles sinensis, ecological niche modeling, biomod2, Chongqing, Plasmodium vivax, Plasmodium falciparum, vector surveillance, spatial epidemiology, China, malaria elimination