One in Three Camels Across Africa and Asia Show Signs of Q Fever Exposure

More than one-third of the camels tested in studies across Africa and Asia show evidence of exposure to Coxiella burnetii, the bacterium behind Q fever, according to a new systematic review and meta-analysis that pooled data from 35 studies conducted on those two continents. The analysis, which covered 11,233 camels tested in studies published from 2000 to late 2025, calculated a pooled prevalence of 33.37 percent, with a 95 percent confidence interval stretching from 23.14 to 45.44 percent. Of the animals tested, 4,363 came back positive for evidence of the pathogen, whether by antibody detection or molecular testing. The finding positions dromedary camels, the dominant livestock species across the arid and semi-arid belts of both continents, as a substantial and previously underappreciated reservoir of one of the most infectious bacteria known to affect humans and livestock alike.

Q fever, also known as coxiellosis, is caused by C. burnetii, an obligate intracellular, Gram-negative bacterium with an exceptionally broad host range spanning humans, domestic livestock, pets, wildlife, reptiles, fish, birds and ticks. Domestic ruminants, including cattle, goats, sheep and camels, serve as the principal reservoirs and sources of human infection. The pathogen’s infectious capacity is remarkable: as few as one to ten organisms can establish an infection in humans through inhalation of contaminated aerosols. Infected animals release the bacterium in substantial concentrations through the placenta, foetal membranes, amniotic fluid, milk, urine and faeces, making animals around the time of birth the major sources of transmission. Compounding the risk, C. burnetii forms spore-like structures that allow it to persist in the environment for extended periods. The first documented case of Q fever in camels dates to the 1950s, and cross-sectional surveys since then have repeatedly flagged evidence of infection in camel populations across Africa and Asia, though limited surveillance has left the true burden poorly characterized.

To close that gap, the research team followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 checklist and applied the condition, context and population framework to structure the search. They queried PubMed, Scopus, Web of Science, Science Direct and Google Scholar for prevalence studies published between 1 January 2000 and 30 November 2025, screening titles and abstracts and then full texts against predefined eligibility criteria requiring a specified sample size, a count of positive animals and a described diagnostic method. Two independent teams carried out the screening, with disagreements resolved through discussion and consensus. Study quality was evaluated by three reviewers using the Joanna Briggs Institute critical appraisal tool for prevalence studies, a nine-item checklist scored as high quality for seven to nine points, moderate for four to six and low for zero to three. Every included study achieved moderate to high methodological quality, so none were excluded on design grounds, and the evidence base was deemed sufficiently robust for quantitative synthesis.

The statistical synthesis revealed extraordinary variability between studies. Cochran’s Q test returned a value of 1278.56 with 34 degrees of freedom, and the I-squared statistic reached 97.3 percent, meaning nearly all of the observed variation was attributable to genuine differences between studies rather than chance. A wide prediction interval of 3.55 to 87.20 percent underscored that point: a future study conducted under similar conditions could plausibly report prevalence anywhere from nearly zero to nearly nine in ten animals. Because of this substantial heterogeneity, the researchers applied a random-effects model. They analysed prevalence proportions on the logit scale to stabilize variances, applied a continuity correction of 0.5 for studies reporting zero positive events, and confirmed robustness with a sensitivity analysis using the Freeman-Tukey double-arcsine transformation, which yielded a nearly identical pooled estimate of 33.90 percent.

Geographically, eligible data came exclusively from Africa and Asia, with 65.7 percent of studies from Africa and 34.3 percent from Asia. African studies spanned seven countries, led by Kenya with seven studies, followed by Egypt with five, Ethiopia with four, Algeria with three, Tunisia with two, and Chad and Sudan with one each. Asian coverage included Iran with four studies, Saudi Arabia with three, Pakistan with two, and single studies from the United Arab Emirates, Jordan and Iraq. The pooled prevalence was numerically higher in Africa at 37.17 percent than in Asia at 26.87 percent, but the difference was not statistically significant, indicating no evidence of a true continental divide. Between countries, however, the variation was stark and statistically significant, ranging from 8.67 percent in Iran to 80.28 percent in Chad. The authors caution that several countries were represented by only a single study, so these figures reflect the available published evidence rather than true national prevalence.

Diagnostic method emerged as the only moderator with a statistically significant effect. Twenty-six studies used enzyme-linked immunosorbent assays (ELISA), which detect host antibodies against C. burnetii and therefore capture both recent and past exposures, while nine studies used polymerase chain reaction (PCR), which detects bacterial DNA and indicates active infection or bacterial presence at the moment of sampling. The pooled seroprevalence by ELISA was 44.06 percent, compared with just 7.76 percent by PCR, a difference the subgroup analysis confirmed as significant. The discrepancy likely reflects both the much larger ELISA sample pool of 9,903 animals against 1,330 tested by PCR and the biology of the assays themselves: the bacterium may be transient or present at low levels outside acute infection phases, depressing PCR detection rates, whereas antibodies persist long after exposure.

Subgroup analyses by sample size and publication year showed numerical but not statistically significant differences. Studies testing more than 400 camels reported a pooled prevalence of 43.05 percent, versus 30.22 percent for smaller studies, and studies published in 2020 or earlier reported 39.43 percent, versus 28.99 percent for those published after 2020. The authors note that more than half of the included studies, 54.3 percent, appeared in the last five years, suggesting growing attention to the disease. Robustness checks reinforced confidence in the headline figure: a leave-one-out analysis, in which each study was sequentially removed and the pooled estimate recalculated, kept the result within a narrow band of 31.97 to 35.34 percent. Excluding the single most influential study raised the estimate to 35.34 percent, and excluding the four most influential outliers identified through Galbraith radial and Baujat diagnostic plots yielded 36.96 percent, with heterogeneity remaining high throughout. Removing the four moderate-quality studies changed the pooled estimate only marginally, to 32.14 percent. Funnel plot symmetry, Begg’s rank correlation test and Egger’s regression test found no clear evidence of publication bias.

Two risk-factor associations stood out from the subset of studies reporting them. Among 15 studies that recorded the sex of sampled animals, the odds of Q fever in female camels were about 2.22 times higher than in males, with a 95 percent confidence interval of 1.53 to 3.24. This aligns with the pathogen’s strong affinity for reproductive tissues, particularly placental trophoblasts, where it replicates to extraordinary titers in a tissue-specific amplification niche absent in males, and is shed during subsequent births and lactation. Female animals are also typically kept longer in production systems for breeding and milk production, prolonging their exposure to contaminated environments. Separately, six studies reporting tick infestation status showed that camels carrying ticks had significantly higher odds of C. burnetii infection, with an odds ratio of 2.09 and a confidence interval of 1.28 to 3.41, consistent with ticks’ recognized role as reservoirs and potential vectors of the bacterium in pastoral systems where acaricide use is limited.

The authors frame the findings as evidence that Q fever constitutes a significant zoonotic concern in camel-producing regions of Africa and Asia, with breeding-age females serving as the primary amplification reservoirs and environmental shedders within herds. They also stress the limits of the evidence base: only English-language, full-text studies from indexed databases were included, non-indexed and non-English literature was missed, and no eligible prevalence data emerged from any region outside Africa and Asia, leaving open whether camel Q fever is a truly global zoonotic threat or remains primarily endemic to these two continents. The very high between-study heterogeneity means the pooled figure should be read as a summary of the available evidence rather than a universally applicable estimate, and direct comparisons between countries are complicated by differences in diagnostic approaches, sampling strategies and study populations. Strengthening surveillance, standardizing diagnostic approaches and conducting targeted epidemiological investigations with broader geographic coverage, the authors conclude, are the priorities needed to characterize the true burden and distribution of C. burnetii in camel populations.

Subject of Research: Global prevalence of Q fever (Coxiella burnetii infection) in camels

Article Title: Global Prevalence of Q Fever in Camels: A Systematic Review and Meta‐Analysis of Studies From Africa and Asia

Article References: Kallu, S. A., Mengistu, S., Asnake, P. W., Wubaye, A. M., Bezabih, M. K., Mekonen, M. T., Alamneh, S. Z., Elmi, Y. A., Sali, K., Endalamew, S. G., Molla, A. M., & Ambaw, Y. G. (2026). Global Prevalence of Q Fever in Camels: A Systematic Review and Meta‐Analysis of Studies From Africa and Asia. Veterinary Medicine and Science, 12(6), Article e71275. https://doi.org/10.1002/vms3.71275

Image Credits: AI Generated

DOI: 10.1002/vms3.71275

Keywords: Q fever, Coxiella burnetii, camels, zoonotic disease, meta-analysis, seroprevalence, ELISA, PCR, Africa, Asia, livestock, one health