Global spatiotemporal patterns of bovine mastitis pathogens from 2000 to 2025: a meta-analysis for precision dairy health Authors: Kai Yang, Guiqiu Hu, Zimo Pei, Meng Lan, Yu Cao, Xuanting Liu, Juxiong Liu, Dewei He, Wenjin Guo, Shoupeng Fu Published: 2026 ABSTRACT AbstractBackgroundBovine mastitis is the most prevalent and economically damaging disease affecting dairy production worldwide. The disease can reduce milk yield and quality, shorten productive herd life, and compromise animal welfare and farm profitability. Although the spectra of responsible pathogens vary widely across regions, climates, and farming systems, long-term global spatiotemporal data that could guide mastitis resistance breeding and precision nutritional strategies remain limited. This meta-analysis aimed to characterize the spatiotemporal dynamics of bovine mastitis pathogens and to evaluate their associations with geography, climate, cattle breed, and husbandry systems.ResultsA total of 137 studies published between 2000 and 2025 were included, covering 37 countries and 208,510 dairy cows. The pooled prevalence estimates and 95% confidence intervals for the five dominant pathogens were as follows: Staphylococcus aureus, 19.59% (95% CI: 17.00%–22.47%); Streptococcus spp., 17.29% (95% CI: 15.14%–19.66%); Escherichia coli, 13.86% (95% CI: 11.22%–17.00%); Coagulase-negative staphylococci, 15.49% (95% CI: 10.12%–22.99%); and Klebsiella spp., 9.92% (95% CI: 7.38%–13.20%). Overall, the pathogen spectrum clearly changed temporally over the 25-year period, with Klebsiella spp. exhibiting the most pronounced upward signal. Subgroup analyses suggested pathogen-specific variability across geographies and climates. The highest prevalence of S. aureus was observed in plateau climates and Africa. At the continent level, Streptococcus spp. were more prevalent in Europe and Asia, whereas climactically, they were more prevalent in plateau and tropical climates. Coagulase-negative staphylococci had the highest continental prevalence in Europe, while climate-specific estimates of its prevalence were relatively high in tropical and subtropical zones.ConclusionsThis study provides long-term global evidence of the substantial temporal and geographic variability of the patterns of various bovine mastitis pathogens, including pathogen-specific evidence for geographic and climatic differences. These spatiotemporal patterns provide important etiological evidence for the targeted genetic improvement of mastitis resistance and pathogen-specific nutritional interventions. These findings provide essential etiological evidence for integrating multidisciplinary technologies including animal breeding, animal nutrition and farm health management to improve bovine mammary health, raise production efficiency and advance the sustainable development of dairy farming.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s40104-026-01488-7.Keywords: Bovine mastitis, Disease resistance breeding, Meta-analysis, Pathogen spectrum, Precision nutrition, Spatiotemporal analysis Introduction Bovine mastitis is the most prevalent disease in the dairy industry worldwide and is widely recognized as one of the most economically important limiters of dairy production. The disease reduces milk yield and quality, increases culling rates, shortens the productive lifespan of dairy cows, and causes substantial economic losses to dairy farms while negatively affecting animal welfare and public health [1–3]. With the massive increases in dairy production in recent years, mastitis prevention and control have become key steps for improving production efficiency and encouraging healthy dairy farming. According to recent market surveys and forecasts, the global costs associated with bovine mastitis, including prevention, diagnosis, and treatment, is estimated to reach approximately $1.5 billion in 2025 and is projected to increase to $2.3 billion by 2032 [4]. Bovine mastitis can be classified as clinical or subclinical according to its clinical presentation. Clinical mastitis is further categorized into grades I–III and is characterized by local or systemic signs of varying intensity, including redness, swelling, heat, and pain of the udder, as well as changes in the physicochemical properties of the produced milk. Subclinical mastitis, in contrast, lacks obvious clinical signs but is associated with an elevated somatic cell count in the milk, making it particularly detrimental to dairy production [5]. Bovine mastitis can be caused by a wide range of pathogens [6], but the dominant pathogens vary substantially across regions, climates, husbandry systems, and management practices [7, 8]. Clarifying the temporal changes in the distributions of these pathogens is essential for developing targeted strategies for mastitis resistance breeding and nutritional regulation. In some regions, mastitis prevention and control still lack sufficient etiological support, limiting the precision and effectiveness of relevant intervention measures. Therefore, characterizing the global spatiotemporal changes in mastitis pathogens is important for guiding healthy dairy farming and delivering precision mastitis control. Mastitis resistance is an important economic trait in dairy cattle breeding; in this context, differences in responsible pathogen spectra may directly affect the selection of resistance-related genes and the direction of breeding programs [9, 10]. Nutritional regulation is an important nonantibiotic strategy for improving mammary gland health and reducing the incidence of mastitis; to achieve these goals, however, different pathogens may require targeted nutritional interventions [11–14]. However, the lack of long-term global pathogen data limits the identification of mastitis resistance loci and the development of precision nutritional formulas for dairy cows. Over the past 25 years, the global dairy industry has undergone substantial development, with the increasing adoption of automated milking systems and standardized health management [15]. These changes have led to greater requirements for mastitis prevention and control. However, previous studies have often been limited in scope, investigating only regional, short-term, and cross-sectional data, and a global, longitudinal analysis covering pathogen information over longer periods remains lacking. In the present study, we collected global data on bovine mastitis pathogens from 137 studies covering 37 countries and 208,510 dairy cows published from 2000 to 2025. The prevalence of 15 pathogens was analyzed, and differences across continents, countries/regions, climates, sampling periods, cow breeds, and farming systems were evaluated. A random-effects meta-analysis model was used to characterize the temporal and spatial distributions of the pathogens and the factors that potentially influence these distributions. The aim was to clarify the global spatiotemporal pattern of mastitis pathogens and to provide a rational basis for developing mastitis resistance breeding, precision nutrition, and healthy dairy farming strategies. Materials and methods Search strategy This study was conducted in accordance with the 2020 PRISMA statement [16] (the checklist can be found in Table S1). To identify eligible studies related to the pathogen spectrum of dairy cow mastitis, the PubMed, Web of Science, and Embase databases were searched using predefined search terms (Table S2). The search was restricted to English-language publications, and no restrictions on publication type were imposed. Screening process Two authors independently screened the titles and abstracts of the studies retrieved from the database searches. The articles considered eligible on this basis were then evaluated according to predetermined inclusion and exclusion criteria (Table S4). At this stage, eligible studies were epidemiological or original research articles that involved sampling after 2000, had clear research objectives, and reported their detection methods and results for pathogenic microorganisms associated with dairy cow mastitis. The exclusion criteria were duplicate publications; non-English articles; reviews; meta-analyses; studies with unclear pathogen information, unavailable data or missing geographic information, unavailable full text, or sampling performed before 2000; case reports; and laboratory-based molecular mechanism studies. The screening decisions were finalized through weekly discussions among the authors. The final included studies are listed in Table S3. Data extraction A predesigned data-extraction template was used to ensure consistency. The extracted variables included first author, title, study retrieval code and database, publication year, sampling year, sampling country, detection method, breed purity, and farming output. Because most studies did not explicitly report the relevant climate zone or continent, these variables were determined according to the reported sampling locations. For studies in which the sampling location was reported only at the national level and spanned multiple climate zones, climate data were not included. Quality assessment Two authors independently assessed the risk of bias according to the methods in previous epidemiological meta-analyses and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [17, 18]. Briefly, the assessment included sample selection, sample size, outcome assessment, and analytical method (Tables S5 and S6). To further evaluate study quality, a checklist developed on the basis of previous studies [18, 19] was used. The checklist contained 10 items, each scored as 1 point for “Yes” or 0 points for “No”. The final score, X, was calculated as the sum of the 10 item scores divided by 10, according to which the study was classified as high quality (X ≥ 0.78), moderate quality (0.44 < X < 0.78), or low quality (X ≤ 0.44). In accordance with the Cochrane Handbook and current statistical recommendations for meta-analyses, quality scores were not directly used as weights in the quantitative synthesis; instead, they were used only as a reference for evaluating the quality of the included study (Tables S7 and S8). The results of the quality assessment was also confirmed through weekly communication among all the authors. Statistical analysis All analyses were performed using R software (version 4.5.2) and RStudio (version 2025.09.2 + 418). Publication trends for pathogen-specific mastitis studies were analyzed using generalized linear models constructed with a negative binomial distribution to account for count data and possible overdispersion. The binom package was used to calculate 95% confidence intervals (CIs) for pathogen-positivity rates. Given the substantial heterogeneity arising from the broad temporal range (2000–2025) and wide geographic distribution of the included studies, meta-analyses of the five major pathogen categories identified in the broader meta-analysis were conducted using a random-effects model in the meta package. Pooled prevalence was estimated using an inverse-variance random-effects model with logit transformation (sm = “PLOGIT”). The DerSimonian‒Laird estimator was used as the primary estimator of τ2, while REML, Paule‒Mandel, and Hartung‒Knapp-adjusted models were used in the sensitivity analyses [20–24]. Heterogeneity was assessed using the Chi-square test (Cochrane’s Q statistic) and the I2 statistic; for the latter, values of 25%, 50%, and 75% were considered indicative of low, moderate, and high heterogeneity, respectively, and P < 0.05 was considered indicative of statistical significance. Heterogeneity and potential publication bias were evaluated using forest plots, funnel plots, Egger’s test, and Begg’s test. A P < 0.05 in either Egger’s test or Begg’s test was considered indicative of potential publication bias. Temporal trends in pathogen-positivity rates from 2000 to 2025 were explored using meta-regression or descriptive trend analysis, depending on the availability and continuity of the data. Subgroup analyses based on continent, country/region, climate, breed purity, and farming system were performed to clarify the effects of genetic and environmental factors on the prevalence of the pathogens. Result stability was evaluated using leave-one-out sensitivity analysis [25]. Analyses restricted to high-quality studies were also conducted [26]. Results Study selection The initial searches across all three databases yielded 13,948 studies. After duplicate records were removed, 8,625 articles remained. After the titles and abstracts were read, 7,576 of these articles were excluded. A further 912 studies were excluded if they had unavailable full text (102), were case reports (1), had a noncompliant sampling time (19), were nonresearch articles (56), were published in a language other than English (2), did not define the pathogens involved (112), did not report the number of sampled dairy cows (55), did not report the number of mastitis cases (25), and had missing or inaccessible epidemiological data (540). Ultimately, 137 studies were eligible for the epidemiological meta-analysis and were included in the final analysis (Fig. 1 and Table S3). Flow chart of study selection Characteristics of the included studies The 137 included studies were conducted across 37 countries, indicating a relatively broad global coverage of studies on bovine mastitis pathogens (Fig. 2A). The number of eligible studies increased after 2010; 56.9% were published during the most recent five years, and 29.9% were published between 2016 and 2020 (Fig. 2B). Most studies were conducted in Asia, whereas South America and Oceania were underrepresented (Fig. 2D). The quality assessment revealed that 74.5% of the included studies were of high quality (Fig. 2C). In terms of pathogen-specific research volume, S. aureus was the most frequently investigated pathogen category, followed by Streptococcus spp., E. coli, coagulase-negative staphylococci (CNS), and Klebsiella spp. (Fig. 2E). Characteristics of the included studies. A Distribution of the included studies by country. B Distribution of publication dates of the included studies. C Quality assessment of the included studies. D Distribution of the included studies by continent. E Number of studies on the included pathogens. F–T Annual changes in the volume of published studies on the epidemiology of pathogens causing bovine mastitis. The dashed line and shaded area represent the regression line fitted using a generalized linear model with a negative binomial distribution and a log link function. The blue line indicates the expected trend based on total publication volume, which is used to determine whether the publication rate for a specific pathogen is faster or slower than the overall publication rate for bovine mastitis Longitudinal analysis of the publication trends from 2000 to 2025 revealed a strong correspondence between research on specific pathogens and the overall growth of the dairy mastitis field. The total number of studies on the epidemiology of major mastitis-causing pathogens has steadily increased. The global publication growth rate (shown as a blue line in the relevant figures) was used as a benchmark to evaluate whether research interest in each specific pathogen (shown as a red line in the relevant figures) was consistent with, exceeded, or lagged behind the overall growth of bovine mastitis research (Fig. 2F–T). For the five major pathogens (Fig. 2F–J), which represented the dominant pathogen categories, the publication trajectories were nearly aligned with the expected baseline, indicating that the research output for these pathogens increased proportionally with that of the field as a whole. Some minor pathogens (Fig. 2K–N) were associated with substantially lower publication volumes and were categorized as low-attention or sporadically reported pathogens; although their publication trends generally aligned with expectations, they appeared to fluctuate more or were relatively flat because of their small baseline size. Other minor pathogens (Fig. 2O–T) consistently had low absolute publication numbers, generally remaining near the baseline, and were associated with wider confidence intervals because of the sparsity of the relevant data. Global pooled prevalence estimates, bias assessments, and robustness analyses for pathogens that cause mastitis in dairy cows Random effects pooled analysis of the studies conducted between 2000 and 2025 revealed that the prevalence of bovine mastitis caused by the five dominant pathogen categories was 19.59% for S. aureus (95% CI: 17.00%−22.47%), 17.29% for Streptococcus spp. (95% CI: 15.14%−19.66%), 13.86% for E. coli (95% CI: 11.22%−17.00%), 15.49% for CNS (95% CI: 10.12%−22.99%), and 9.92% for Klebsiella spp. (95% CI: 7.38%−13.20%) (Figs. 3A, 3C, 4A, 4C, and 4E; Tables 1, 2, 3, 4, and 5). Substantial heterogeneity was observed across all five pathogen categories, with I2 values ranging from 98.1% to 99.6% (all P < 0.001). The wide confidence intervals indicated that the true prevalence may vary considerably across future or local study settings, reflecting differences in geographic region, sampling strategy, diagnostic method, and host population characteristics. Meta-analysis of major pathogens causing dairy cow mastitis (2000–2025). A and B Forest plot of the results of the meta-analysis and leave-one-out sensitivity analysis, respectively, for Staphylococcus aureus. C and D Forest plot of the results of the meta-analysis and leave-one-out sensitivity analysis, respectively, for Streptococcus spp. Meta-analysis of major pathogens causing dairy cow mastitis (2000–2025). A and B Forest plot of the results of the meta-analysis and leave-one-out sensitivity analysis, respectively, for Escherichia coli. C and D Forest plot of the results of the meta-analysis and leave-one-out sensitivity analysis, respectively, for coagulase-negative staphylococci. E and F Forest plot of the results of the meta-analysis and leave-one-out sensitivity analysis, respectively, for Klebsiella spp. Table 1S. aureus subgroup informationVariable (Subgroups)Number of study (n)Number sampled (n)Number of positive (n)Rate of positive, %95% CI lower, %95% CI upper, %Heterogeneity Chi-squared (χ2)I2, %P-value for heterogeneityContinent5,353.8899.9< 0.0001 North America863,6596,81511.729.5214.34125.0494.4< 0.0001 Oceania33,6584725.161.2918.535.7594.4< 0.0001 Africa192,91067327.0418.1938.18443.3395.9< 0.0001 South America1498265.223.597.54--- Europe1467,61617,53516.5413.3520.311,322.2999.3< 0.0001 Asia3526,8663,88219.5917.0022.472,488.8798.6< 0.0001Country9,514.0899.7< 0.0001 United States357,7746,1307.84.1714.1543.9695.5< 0.0001 Canada45,78065212.079.1915.6936.9291.9< 0.0001 Mexico11053331.4323.3440.83--- Australia33,6584725.161.2918.535.7594.4< 0.0001 Algeria2393215.343.518.06000.9724 Egypt557017744.6124.766.4259.4293.3< 0.0001 Ethiopia81,00241539.929.0951.888.5392.1< 0.0001 Tunisia3913465.251.6115.830.3393.4< 0.0001 South Africa1321443.7528.1760.67--- Brazil1498265.223.597.54--- Kosovo11522717.7612.5024.61--- Germany33,9862074.482.96.868.6476.80.0133 France11,210978.026.629.68--- Finland129,9697,91726.4225.9226.92--- Bosnia and Herzegovina11205949.1740.3958--- Romania132515046.1540.8151.59--- Poland21483725.5917.8235.281.4731.80.2258 Norway130,1548,91929.5829.0730.10--- United Kingdom1929202.151.403.30--- Hungary11014948.513958.14--- Estonia15225310.157.8513.04--- India61,46529222.8610.8241.99164.1297< 0.0001 China1322,4802,43614.4211.7517.57245.8995.1< 0.0001 Iran25077414.867.2328.0910.3490.30.0013 Bangladesh32788136.943.3490.8572.2197.2< 0.0001 Nepal23256119.1512.0329.13.85740.0497 Thailand18444.761.8711.61--- Pakistan51,22663341.7924.6561.17143.9497.2< 0.0001 Malaysia1743952.7041.4863.66--- Japan1572442.1130.1955.02--- Sri Lanka137023864.3259.3269.04---Climate35.7594.4< 0.0001 Plateau climate680829435.0027.8542.9124.7279.80.0002 Tropical climate122,45872131.5918.5448.36456.7297.6< 0.0001 Temperate climate2990,82619,98714.8712.0318.243,971.2899.3< 0.0001 Subtropical climate2261,2397,17120.4512.7331.171935.198.9< 0.0001Breeding type595.9599.8< 0.0001 Purebred dairy cows934,3158,50821.9514.3232.12531.6198.5< 0.0001 Crossbred dairy cows127,58987415.19.7722.61381.7597.1< 0.0001Production system7.0785.80.0079 Intensive2589,69219,9772214.332.13,853.4899.4< 0.0001 Extensive13205540727.317.839.4220.8194.6< 0.0001 S. aureus subgroup information Table 2Streptococcus spp. subgroup informationVariable (Subgroups)Number of study (n)Number sampled (n)Number of positive (n)Rate of positive, %95% CI lower, %95% CI upper, %Heterogeneity Chi-squared (χ2)I2, %P-value for heterogeneityContinent4,211.9899.9< 0.0001 North America562,43021,11216.638.0531.221,086.3399.6< 0.0001 Oceania33,6581,3678.510.7653.05120.1298.3< 0.0001 Africa9166322814.589.4321.8876.7289.6< 0.0001 Europe1470,58914,82220.0716.3424.401,405.5699.1< 0.0001 Asia2927,1724,77917.2613.5621.721,375.1498< 0.0001Country6,181.999.6< 0.0001 Ethiopia33608523.6219.5128.280.0800.9622 Algeria122652.210.955.07--- Egypt13356920.6016.6125.25--- Estonia152215028.7425.0232.76--- Australia33,6581,3678.510.8653.05120.1298.3< 0.0001 Poland21486332.486.377.4919.4294.8< 0.0001 Germany33,98696223.1317.829.4925.3692.1< 0.0001 France11,21015112.4810.7414.46--- Finland129,9694,70715.7115.3016.12--- Canada45,78075413.358.1521.10130.3497.7< 0.0001 Kosovo1152127.894.5713.29--- United States156,65020,35835.9435.5436.33--- Bangladesh443411222.4313.5134.8719.2884.40.0002 South Africa1321237.5022.9354.75--- Norway130,1548,09626.8526.3527.35--- Serbia18178.644.2516.78--- Sri Lanka1370123.241.865.58--- Thailand226214864.5437.9679.3610.9190.80.001 Tunisia2613442.960.2032.197.72870.0055 Uganda1971313.40821.59--- Hungary11013231.6823.4241.29--- Iran11803117.2212.4123.41--- India23,995411.030.761.400.4500.5003 United Kingdom24,26664214.4812.1317.203.873.70.0514 China1921,9314,43517.9214.3122.21706.9697.5< 0.0001Climate1,375.1498< 0.0001 Plateau climate33608523.619.528.30.0800.9622 Tropical climate101,57037322.6912.6437.31200.896< 0.0001 Temperate climate3091,50118,55417.8115.3920.522,214.8598.7< 0.0001 Subtropical climate1262,91820,6927.733.5715.911,009.1298.9< 0.0001Breeding type44.5497.8< 0.0001 Purebred dairy cows534,3225,37419.8813.7527.86232.4898.3< 0.0001 Crossbred dairy cows106,2431,18920.5113.2130.43335.2997.3< 0.0001Production system5.3381.20.021 Intensive2189,13818,03416.6213.8719.792,206.8499.1< 0.0001 Extensive71,29322814.335.5332.25200.6497< 0.0001 Streptococcus spp. subgroup information Table 3E. coli subgroup informationVariable (Subgroups)Number of study (n)Number sampled (n)Number of positive (n)Rate of positive, %95% CI lower, %95% CI upper, %Heterogeneity Chi-squared (χ2)I2, %P-value for heterogeneityContinent1,297.599.6< 0.0001 North America66,57471111.427.6716.69131.0496.2< 0.0001 Oceania23,3702204.791.9711.203.6172.30.0574 Africa8166324711.057.0216.9966.2989.4< 0.0001 South America1257145.453.278.93--- Europe1062,9614,29214.267.5325.382,890.5699.7< 0.0001 Asia2522,3663,27117.4914.5320.92534.0695.5< 0.0001Country5,353.8898.9< 0.0001 Ethiopia3360298.125.7011.440.5200.7704 Thailand18411.190.216.44--- Algeria12263314.6010.5919.80--- Egypt14325913.6610.7417.22--- Australia23,3702204.791.9711.23.6172.30.0574 Pakistan11111614.419.0722.14--- Brazil1257145.453.278.93--- Poland21482717.7910.8827.731.54350.2149 Bosnia and Herzegovina112032.500.857.09--- Germany12498634.5428.9140.64--- France11,21065153.8050.9856.59--- Finland129,9691,4684.904.665.15--- Canada45,78054010.007.8412.6724.1187.6< 0.0001 United States174116722.5419.6725.68--- Bangladesh446014327.1111.5751.3958.1294.8< 0.0001 Mexico15347.552.9717.86--- South Africa132515.626.8631.75--- Norway130,15418596.175.906.44--- Serbia18178.644.2516.78--- Sri Lanka1370267.034.8410.10--- Tunisia26131217.970.4960.2724.3495.9< 0.0001 Hungary11012019.8013.2028.62--- Iraq150183624.1449.86--- Iran293214015.0612.9017.510.9900.3204 India150142817.4741.67--- United Kingdom192917118.4116.0521.03--- Jordan13807419.4715.8123.75--- Vietnam1686507.295.579.48--- China1219,2432,78917.1113.7021.16275.9496< 0.0001Climate339.6899.1< 0.0001 Plateau climate3360298.125.7011.440.5200.7704 Tropical climate8160314612.126.2622.1796.4892.7< 0.0001 Temperate climate2483,6957,02613.759.9718.673,755.0599.4< 0.0001 Subtropical climate9184939119.1912.1329.02218.2895.9< 0.0001 Breeding type852.1299.9< 0.0001 Purebred dairy cows533,922187612.306.5122.04371.4198.9< 0.0001 Crossbred dairy cows127,5641,18516.1311.8021.67258.5395.7< 0.0001Production system195.3999.5< 0.0001 Intensive1581,4216,91016.3911.0523.633,704.2799.6< 0.0001 Extensive12183133112.537.4820.24177.9593.8< 0.0001 E. coli subgroup information Table 4CNS subgroup informationVariable (Subgroups)Number of study (n)Number sampled (n)Number of positive (n)Rate of positive, %95% CI lower, %95% CI upper, %Heterogeneity Chi-squared (χ2)I2, %P-value for heterogeneityContinent6,555.1499.9< 0.0001 North America45,50850210.294.7021.05228.0398.9< 0.0001 Oceania23,507542.790.1338.61115.9899.1< 0.0001 Africa10181344520.8211.2935.20306.697.1< 0.0001 Europe533,98113,79822.075.9855.761,330.7399.7< 0.0001 Asia1119,025181315.499.6224.00924.0598.9< 0.0001Country8,851.1799.8< 0.0001 Ethiopia33605013.9710.7517.970.8700.6482 Thailand1843238.1028.4548.78--- Algeria239323760.2955.3665.020.600.4392 Egypt150483.1518.84--- Australia23,507542.790.1338.61115.9899.1< 0.0001 Poland1483164.5850.4476.57--- Germany12,8831725.975.166.89--- Finland129,96913,50845.0744.5145.64--- Canada34,76745712.154.7627.70208.6799< 0.0001 Kosovo11525838.1630.8246.08--- United States1741456.074.578.03--- Tunisia391313315.287.0030.1840.6195.1< 0.0001 Uganda1972121.6514.6230.84--- India25647612.910.9669.3677.6498.7< 0.0001 United Kingdom1929293.122.184.45--- China818,377170514.218.3323.19775.1199.1< 0.0001Climate144.7997.9< 0.0001 Plateau climate33605013.9710.7517.970.8700.6482 Tropical climate474512919.947.2444.2786.7396.5< 0.0001 Temperate climate1352,39015,19716.457.5932.086,723.6599.8< 0.0001 Subtropical climate7194740119.359.2036.23289.7997.9< 0.0001Breeding type1,477.9399.9< 0.0001 Purebred dairy cows533,83313,84021.186.7250.051,208.6299.7< 0.0001 Crossbred dairy cows65,25154116.128.1929.2918897.3< 0.0001Production system40.4997.5< 0.0001 Intensive950,56115,08715.615.3330.476,258.2499.9< 0.0001 Extensive61,26127119.217.7040.39221.1997.7< 0.0001 CNS subgroup information Table 5Klebsiella spp. subgroup informationVariable (Subgroups)Number of study (n)Number sampled (n)Number of positive (n)Rate of positive, %95% CI lower, %95% CI upper, %Heterogeneity Chi-squared (χ2)I2, %P-value for heterogeneityContinent249.2598< 0.0001 North America55,5612163.442.135.5231.8987.5< 0.0001 Oceania128810.350.011.92--- Africa2524798.001.2637.147.3486.40.0067 South America125772.721.335.51--- Europe3430294.291.2413.778.2175.60.0165 Asia1626,7912,72716.8712.0423.141,119.9298.7< 0.0001Country343.796.2< 0.0001 Ethiopia17222.780.779.57--- Thailand1841011.906.6020.54--- Australia128810.350.061.94--- Brazil125772.721.335.51--- Poland1100331.038.45--- Germany12492510.046.8914.40--- Canada34,7671632.641.464.7114.7486.40.0006 United States1741516.885.278.94--- Mexico15323.771.0412.75--- Serbia18111.230.226.67--- Tunisia14527717.0413.8520.78--- India2180158.395.1213.450.2500.6168 Vietnam146810322.0118.4925.98--- China1226,0592,59918.3812.5026.191,073.5399< 0.0001Climate189.7298.4< 0.0001 Plateau climate17222.780.779.57--- Tropical climate61,0421378.353.8517.1650.8590.2< 0.0001 Temperate climate1427,4822,2029.035.9613.46941.2598.6< 0.0001 Subtropical climate41,10421319.199.0436.1866.4795.5< 0.0001Breeding type25.3796.1< 0.0001 Purebred dairy cows23,31718913.101.2963.53181.8199.4< 0.0001 Crossbred dairy cows1012,3891,0358.274.6714.24559.5998.4< 0.0001Production system11.2891.10.0008 Intensive1627,5692,70013.189.2018.5212.1183.50.0023 Extensive3577816.531.5923.221,087.8298.6< 0.0001 Klebsiella spp. subgroup information The results of Egger’s tests and Begg’s tests, together with visual inspection of the funnel plots, suggested potential publication bias or small-study effects for some pathogen categories (Figs. S1–S5 and Table S10). Specifically, no significant publication bias was detected for S. aureus by either Egger’s test or Begg’s test (P > 0.05), although its funnel plot contains some visual asymmetry. For Streptococcus spp., E. coli, and CNS, the significance of the results of Egger’s and/or Begg’s test and the asymmetric funnel plots indicated possible publication bias or small-study effects. To evaluate the stability of the pooled estimates, leave-one-out sensitivity analysis was conducted by excluding one study at a time (Figs. 3B, 3D, 4B, 4D, and 4F). After the removal of the individual studies, the recalculated pooled prevalence estimates for all five dominant pathogen categories remained close to the original estimates; the nearly vertical distribution of the estimates in the sensitivity plots indicated that no single study disproportionately influenced the pooled results, supporting the robustness of the global summary estimates. To further evaluate the robustness of the pooled estimates under conditions of substantial heterogeneity, alternative random-effects specifications were compared for the five dominant pathogen categories (Table S11). Across the DerSimonian–Laird models, REML models, Paule–Mandel models, and REML models with Hartung–Knapp adjustment, the pooled prevalence estimates remained broadly consistent. Specifically, the estimates ranged from 19.52% to 19.59% for S. aureus; from 16.42% to 17.29% for Streptococcus spp.; from 13.81% to 13.86% for E. coli; from 15.49% to 15.50% for CNS; and from 9.32% to 9.92% for Klebsiella spp. Although the confidence intervals widened under some alternative specifications, particularly after Hartung–Knapp adjustment, the overall ranking and interpretation of the dominant pathogen categories did not change markedly. Sensitivity analyses restricted to high-quality studies yielded results similar to those obtained from the broader analysis (Table S12). After the analysis was restricted to high-quality studies, the pooled prevalence estimates were 20.92% for S. aureus, 18.53% for Streptococcus spp., 12.67% for E. coli, 16.31% for CNS, and 10.80% for Klebsiella spp. These estimates are similar to the corresponding estimates from the full dataset, indicating that the main findings were not dominated by lower-quality studies. However, the heterogeneity remained substantial in the high-quality-only analyses, with I2 values ranging from 97.7% to 98.9%, suggesting that between-study variability was attributable mainly to true differences in geographic region, diagnostic method, study population, and production system rather than to study quality alone. Global prevalence and spatiotemporal variability of major mastitis pathogens in dairy cows Geographic analysis revealed significant differences among the five dominant pathogen categories (all P < 0.001) (Tables 1, 2, 3, 4, and 5). The corresponding numbers of continents and countries in these analyses were 6 and 31 for S. aureus, 5 and 25 for Streptococcus spp., 6 and 29 for E. coli, 5 and 16 for CNS, and 6 and 14 for Klebsiella spp., respectively (Fig. 5). Further analysis revealed clear pathogen-specific distribution patterns. S. aureus had the highest prevalence in Africa and Asia, with pooled prevalence values of 27.04% and 19.59%, respectively (Fig. 5A). Streptococcus spp. were most prevalent in Europe and Asia, with pooled prevalence values of 20.07% and 17.26%, respectively (Fig. 5B). E. coli was most prevalent in Asia and Europe, with pooled prevalence values of 17.49% and 14.26%, respectively (Fig. 5C). CNS was most prevalent in Europe, whereas Klebsiella spp. were most prevalent in Asia (Fig. 5G and J). Detailed continental estimates and confidence intervals are provided in Tables 1, 2, 3, 4, and 5. At the country level, hotspot regions also differed among the pathogen categories. S. aureus showed evidence of high prevalence in Sri Lanka and Malaysia, with country-level estimates exceeding 50%. Streptococcus spp. were most common in Thailand and South Africa, whereas E. coli had the highest country-level prevalence in France. CNS showed the highest prevalence values in Poland and Algeria, and Klebsiella spp. were most prominent in Vietnam. Spatiotemporal distribution of major pathogens causing dairy cow mastitis (2000–2025). A Country-level ranked prevalence plots for Staphylococcus aureus. B Country-level ranked prevalence plots for Streptococcus spp. C Country-level ranked prevalence plots for Escherichia coli. D Temporal trend in the prevalence of Staphylococcus aureus in causing dairy cow mastitis. E Temporal trend in the prevalence of Streptococcus spp. in causing dairy cow mastitis. F Temporal trend in the incidence of E. coli in causing dairy cow mastitis. G Country-level ranked prevalence plots for coagulase-negative staphylococci. H Temporal trend in the prevalence of coagulase-negative staphylococci in causing dairy cow mastitis. I Temporal trend in the prevalence of Klebsiella spp. in causing dairy cow mastitis. J Country-level ranked prevalence plots for Klebsiella spp. Meta-regression analysis was used to evaluate temporal trends in the prevalence of the major bovine mastitis pathogens from 2000 to 2025 (Fig. 5D, E, F, H, and I). S. aureus showed a fluctuating but generally increasing pattern across sequential 5-year intervals. The prevalence of Streptococcus spp. increased from 2000 to 2020 but declined from 2021 to 2025. E. coli began to increase in prevalence after 2006, reaching 31.7% from 2021 to 2025, indicating an overall increasing trend. The prevalence of CNS peaked at 44.3% during 2006–2010 and then tended to stabilize. Among the five dominant pathogen categories, Klebsiella spp. showed the most pronounced upward temporal trajectory. In all the regression plots in the figures above, the bubble size represents the inverse-variance weight of each study. Period-stratified geographic analysis revealed that the countries with the highest reported prevalence values changed over time, indicating that pathogen hotspots were dynamic rather than fixed over the 25-year period (Figs. S6–S10). For example, the leading hotspot regions for S. aureus, E. coli, CNS, and Klebsiella spp. shifted across successive periods, whereas Streptococcus spp. repeatedly demonstrated high prevalence values in North America during later periods. These results suggest that temporal changes in bovine mastitis pathogen patterns were accompanied by geographic redistributions of high-prevalence regions. Detailed period-specific hotspot countries and prevalence estimates are provided in Figs. S6–S10. Sources of heterogeneity in the prevalence of major pathogens causing mastitis in dairy cows From 2000 to 2025, substantial heterogeneity was observed in the pooled prevalence estimates for all five dominant categories of pathogen causing bovine mastitis. To explore potential sources of heterogeneity, subgroup analyses were conducted according to continent, climate type, breed purity, and farming system (Tables 1, 2, 3, 4, and 5 and Figs. S11–S35). The results revealed that subgroup differences were pathogen specific rather than consistent for all predefined variables. For the continent-level comparisons, significant between-subgroup differences were observed for S. aureus, E. coli, and Klebsiella spp. but not for Streptococcus spp. and CNS. Climate-related subgroup differences were significant for S. aureus, Streptococcus spp., and E. coli but not for CNS or Klebsiella spp. In contrast, breed purity did not significantly differ between the subgroups for any of the five dominant pathogen categories. Farming system showed a significant between subgroup difference only for S. aureus, while no significant differences were observed for Streptococcus spp., E. coli, CNS, or Klebsiella spp. Temporal trends and subgroup analyses of minor pathogens causing dairy cow mastitis Geographic analysis revealed that minor mastitis pathogens exhibited uneven and pathogen-specific regional distribution patterns (Figs. S36–S45). The reported prevalence estimate of Enterococcus spp. was relatively high in Poland, China, Uganda, and Kosovo, whereas that of yeast species was particularly high in one Chinese study, with additional moderate estimates reported in Algeria and parts of Europe. High-prevalence estimates were also detected for Pseudomonas spp. and Bacillus spp., especially in individual studies from China, India, and Bangladesh. Enterobacter spp. were relatively more prevalent in Mexico than in other reported regions. Because several of these estimates were derived from a small number of studies, the validity of these apparent hotspots should be interpreted cautiously and verified further through standardized regional surveillance. Detailed country-level estimates are provided in Figs. S36–S45 and Tables S13–S22. The geographic distribution of several minor pathogens also showed marked regional specificity (Figs. S36–S45). The prevalence of Mycoplasma spp. was higher in Iran (29.57%) and Estonia (17.24%) than in most other reported countries, whereas the prevalence of Proteus spp. was high in Mexico (28.30%). In contrast, Corynebacterium spp. showed a relatively even distribution without clear extreme hotspots, while the prevalence estimates of Trueperella spp. and Lactococcus spp. generally remained low. Meta-regression analysis was used to evaluate the temporal trends in the prevalence estimates of minor mastitis pathogens from 2000 to 2025 (Figs. S36–S45). The prevalence of Enterococcus spp., Pseudomonas spp., yeast, and Bacillus spp. tended to increase overall, with more evident increases for Pseudomonas spp. and Bacillus spp. Corynebacterium spp. also showing a positive temporal trend in its prevalence, although the slope was relatively modest. In contrast, the abundances of Mycoplasma spp. and Trueperella spp. tended to decrease over time. In all the regression plots in the figures above, the bubble size represents the inverse-variance weight of each study. Owing to insufficient continuous time series data, the prevalence estimates of Enterobacter spp., Proteus spp., and Lactococcus spp. could not be robustly analyzed using meta-regression. Therefore, temporal changes in the occurrence of these pathogens were evaluated descriptively using bar charts (Figs. S43B, S44B, and S45B). Enterobacter spp. were reported mainly between 2011 and 2020, whereas Proteus spp. and Lactococcus spp. were reported predominantly after 2016. Because the available data for these pathogens are sparse and temporally discontinuous, the current evidence is insufficient to support reliable long-term trend estimations. Discussion This study provides a comprehensive overview of changes in the global mastitis pathogen spectrum. Overall, the findings of this meta-analysis suggest that dairy cow mastitis remains largely associated with several predominant pathogens, but the distribution of these pathogens is not static. Instead, the pathogen spectrum shows clear temporal shifts and marked spatial heterogeneity. Subgroup analyses suggested that geographic and climatic factors may contribute to this heterogeneity in a pathogen-specific manner. These results highlight the need for region-specific surveillance and targeted prevention strategies rather than a uniform, global control model. In accordance with the PRISMA guidelines, in this study, global evidence on the etiology of dairy cow mastitis from 2000 to 2025 was systematically synthesized. Over the past 25 years, the trajectories for major mastitis pathogens, including S. aureus, Streptococcus spp., E. coli, CNS, and Klebsiella spp., reported in the literature have generally followed the overall growth of mastitis research. This observation suggests that these organisms have remained important within the global mastitis pathogen spectrum and continue to pose challenges for modern dairy production despite the use of long-standing control efforts. In contrast, less frequently reported pathogens appear to occupy a more peripheral but relatively stable position in the literature. Their limited representation may reflect opportunistic pathogenicity, regional clustering, lower detection priority, or inconsistent reporting across studies. Although current evidence does not indicate that these minor pathogens have become a major global research focus, they continue to be detected, suggesting that they should not be overlooked, particularly with regard to herd-specific outbreaks, antimicrobial resistance monitoring, and changes in host susceptibility. Given the broad temporal and geographic scope of this meta-analysis, the substantial between-study heterogeneity was not unexpected. The use of random effects modeling, heterogeneity assessment, publication bias evaluation, and leave-one-out sensitivity analysis helped improve the reliability of the pooled estimates. Although evidence of possible publication bias or small-study effects was observed for several pathogens, the sensitivity analysis revealed that the overall estimates were not influenced disproportionately by any single study. To further evaluate whether the findings depended on a single statistical specification, the results of several random-effects models were evaluated, including the DerSimonian–Laird, restricted maximum likelihood, Paule–Mandel, and Hartung–Knapp-adjusted approaches. The broadly consistent estimates obtained with these models support the stability of the main conclusions under different assumptions for between-study variance estimation and uncertainty adjustment. This multimodel validation is particularly important, given the substantial heterogeneity observed in the global data, because reliance on a single random-effects estimator may lead to overconfidence in or model-dependent interpretations of the results. Nevertheless, such consistency across models does not eliminate the underlying across-study heterogeneity; rather, it indicates that the major conclusions are statistically robust, whereas the wide intervals still indicate that the prevalence of the different pathogens may vary considerably across regions, production systems, and future study settings. Therefore, the main findings regarding temporal changes and spatial variation in the major mastitis pathogen spectrum appear to be statistically stable. The pooled prevalence estimates revealed that S. aureus, Streptococcus spp., CNS, E. coli, and Klebsiella spp. were the main pathogens associated with dairy cow mastitis globally. Among them, S. aureus and Streptococcus spp. were particularly prominent, which is consistent with their established roles as major contagious and environmental mastitis pathogens, respectively [27]. However, the extremely high heterogeneity observed across the studies indicates that these pooled estimates should be interpreted as global reference values rather than fixed benchmarks for specific countries, regions, or farms. The wide confidence intervals further suggest that local pathogen prevalence estimates may deviate substantially from the global mean. Therefore, the global pooled estimates are the most useful for identifying broad pathogen patterns, whereas regionally, control targets should be calibrated on the basis of local surveillance data, diagnostic practices, herd structure, and management conditions. Geographic analysis revealed marked spatial heterogeneity in the distribution of the major mastitis pathogens. These differences may reflect true epidemiological variability involving factors such as climate, herd structure, biosecurity conditions, and production systems, but they may also be shaped by differences in case composition, sampling strategy, diagnostic method, study population, and sample size [28]. Similar caution is needed when interpreting the results for minor pathogens, for which isolated, high estimates in certain regions may be more sensitive to sparse-data or local reporting effects. Therefore, future surveillance efforts should prioritize standardized diagnostic protocols and more balanced geographic coverage to verify apparent hotspots and improve the comparability of pathogen-prevalence estimates across regions. The results of the temporal analysis suggested that the major mastitis pathogen spectrum has not remained fixed but rather has changed over the past 25 years. S. aureus, E. coli, and Klebsiella spp. in particular showed evidence of persistent or increasing abundance, whereas the patterns for Streptococcus spp. and CNS appeared to be more variable over time. These changes may be related to shifts in herd management, housing systems, antimicrobial use, diagnostic capacity, and environmental exposure. However, the period-specific estimates should also be interpreted cautiously because differences in sample size, study weight, and regional data availability may influence the apparent direction of the temporal trends. The increasing prevalence estimates observed for Klebsiella spp. are especially noteworthy because this pathogen is often associated with environmental reservoirs and severe clinical outcomes, suggesting that it should receive greater attention in future mastitis surveillance and control programs. The results of geographical-temporal analysis further suggest that regional, high-risk profiles for mastitis pathogens are dynamic rather than fixed. Apparent shifts in pathogen hotspots over time may reflect changes in dairy industry structure, herd management, diagnostic strategies, antimicrobial use, biosecurity practices, and environmental pressures [29], but they may also be affected by uneven surveillance efforts and differences in the studies available for each period and region. Therefore, mastitis prevention and control strategies should be developed on the basis of regularly updated pathogen profiles rather than isolated cross-sectional evidence or empirical assumptions. Establishing continuous regional surveillance systems could help identify emerging hotspots, allow adjustment of intervention priorities, and improve the timeliness of pathogen-specific control strategies. The temporal changes described above may be partly understood in the context of the structural transformation of the dairy industry. Over the past 25 years, dairy production has become increasingly more intense, with wider adoption of automated milking systems and growing concerns about antimicrobial resistance [30, 31]. Changes in housing conditions, environmental exposure, stocking density, milking procedures, and hygiene management may have altered the relative abundances of environmental pathogens such as E. coli and Klebsiella spp. in some regions. In addition, the changing profiles of less frequently reported pathogens suggests that opportunistic or atypical organisms may become regionally abundant under specific management, environmental, or therapeutic conditions. Such shifts may not only reflect true epidemiological changes but also be associated with improved diagnostic coverage and greater reporting capacity. Therefore, future surveillance systems should not only focus on traditional major pathogens but also include specific minor or emerging pathogens, especially in regions where management practices, antimicrobial use, or environmental conditions are changing rapidly. The observed differences across climate zones suggest that climatic conditions may be important contributors to the global heterogeneity in the distributions of mastitis pathogens. Temperature, humidity, bedding moisture, heat stress, and the barn microenvironment can influence pathogen survival, environmental contamination, teat-end exposure, and infection pressure [32]. However, the effect of climate is not deterministic, as it may be modified by farm management. In warm and humid regions, for example, improved bedding management, ventilation, environmental hygiene, and heat-stress control may help reduce the burden of environmental pathogens. Production systems may shape pathogen distribution. Differences between intensive and extensive systems are particularly important for environmental or opportunistic pathogens, as stocking density, housing condition, manure accumulation, bedding management, and milking hygiene can directly affect exposure pressure. In contrast, contagious or persistent pathogens may be influenced not only by the production system itself but also by herd-carrier status, milking-time transmission, chronic infection management, and culling strategies. The findings of this study indicate that climate- and production-related risks should be considered jointly when region-specific mastitis control programs are designed. Breed structure may also contribute to the observed heterogeneity in mastitis pathogen distribution. Differences between purebred and cross-bred dairy cows should be interpreted cautiously, as breed composition is often closely linked with geography, production intensity, herd size, housing conditions, and management practices [33–35]. High-yielding purebred dairy cows may face greater metabolic and udder health pressure than their cross-bred counterparts, whereas cross-bred dairy cows may be raised under distinct environmental and management conditions. Therefore, the apparent breed-related differences in pathogen prevalence were likely shaped by the combined effects of genetic background, immune response, anatomical traits, and production systems [36–38]. Practically, mastitis prevention efforts should be stratified not only by region, climate, and time period but also by herd structure and farm management. The findings of this study also have implications for mastitis resistance breeding: rather than rely on broad, nonspecific selection for mastitis resistance, future breeding programs should integrate pathogen-specific phenotypes with candidate immune-related genes or SNP markers, such as TLR4 and BoLA, to improve selection precision [39, 40]. In addition, regional differences in pathogen spectra suggest that the breeding objectives may need to be adapted to the local epidemiological condition. Such locally informed cross-breeding may help balance milk production, environmental adaptability, and disease resistance, but such strategies would need to be validated using farm-level pathogen surveillance and genetic performance data. In some countries and regions, limited access to routine bacteriological testing results in a reliance on clinical signs for mastitis treatment decision-making, which may reduce the precision of antimicrobial use [41–43]. The spatiotemporal heterogeneity observed in this study further supports the need for auxiliary strategies tailored to local pathogen profiles. Targeted nutritional regulation strategies may complement hygiene management, diagnostic testing, and antimicrobial stewardship by improving host immunity, maintaining mammary epithelial barrier function, and reducing inflammatory damage. Regarding Gram-positive pathogens such as S. aureus, Streptococcus spp., and CNS, previous studies have suggested that selenium-vitamin E, ω-3 polyunsaturated fatty acids, and probiotics may support antioxidant levels, immune regulation, and epithelial barrier integrity [44–47]. With respect to Gram-negative pathogens such as E. coli and Klebsiella spp., plant essential oils, tannins, and prebiotics have been investigated for their potential roles in altering microbial balance and alleviating endotoxin-related mammary inflammation [48–51]. The development of nutritional strategies may also need to consider climate and production systems, such as heat-stress mitigation in hot and humid regions or mineral and immune-supportive supplementation depending on the management condition. However, these interventions should be regarded as supportive measures rather than substitutes for pathogen identification, environmental control, and rational antimicrobial use. Future mastitis control strategies should combine spatial surveillance with temporal trend monitoring to support regular updates of regional pathogen profiles. Such systems could help identify changes in dominant pathogens early and guide timely adjustments of prevention priorities. In warm and humid regions or for intensive production systems, environmental management, including bedding control, ventilation, hygiene, and heat-stress mitigation, should be strengthened to reduce environmental pathogen pressure. In herds with a high contagious or persistent pathogen burden, milking procedures, teat disinfection, identification of carrier cows, and chronic infection management should be prioritized. Overall, pathogen-specific control programs should be adaptive, informed by local conditions, and regularly updated rather than be based on static global averages. This study has several limitations. First, the included studies were unevenly distributed across regions, with relatively greater representation from research-heavy areas and limited evidence from some continents. This unevenness may affect the representativeness of the global pooled estimates. Second, climate and continental information was not consistently reported in the original studies. Consequently, some subgroup variables had to be inferred from the sampling location, and climate classification was limited when the country covered multiple climate zones, which may have reduced the specificity of the climate-related analyses. Third, differences in sampling strategies, diagnostic methods, study populations, and clinical or subclinical case composition likely contributed to the substantial heterogeneity observed across the studies. These limitations should be considered when interpreting the global estimates and applying the findings to specific regional or farm-level contexts. Overall, however, these results should be interpreted with caution. Differences in study scale, sampling strategy, diagnostic method, reporting practice, and regional surveillance effort may partly explain the observed heterogeneity, and thus the results should not be interpreted entirely as true epidemiological changes [52, 53]. In particular, estimates for less frequently reported pathogens may be more vulnerable to sparse-data and regional reporting bias. Thus, although this study provides a useful reference for understanding mastitis pathogen distributions worldwide, regional prevention and control strategies should still be developed on the basis of local surveillance data and farm-level epidemiological evidence [54]. Future studies should promote standardized case definitions, sampling strategies, and diagnostic procedures to improve comparability across regions and reduce methodological heterogeneity. Greater continuity in surveillance is also needed in underrepresented regions to improve the balance and reliability of the global evidence. Additionally, regional studies should consider investigating pathogens with potentially emerging importance and evaluate their relationships with environmental management, antimicrobial use, and animal resistance profiles. Such efforts would help distinguish true epidemiological changes from differences caused by surveillance effort, diagnostic capacity, or reporting practices. Future studies could use the spatiotemporal pathogen distribution data generated in this study to conduct genome-wide association analyses of pathogen-specific resistance traits, identify key loci associated with resistance to dominant pathogens, and support the development of molecular breeding tools for precision selection. In parallel, future research should further investigate the associations among pathogen-specific mammary damage mechanisms, nutritional regulation, and mammary gland health, with the goal of developing regional and pathogen-specific nutritional strategies for dairy cows and improving combined mastitis prevention and control systems. Conclusions The results revealed substantial temporal and spatial worldwide variability in bovine mastitis pathogens. Subgroup analyses similarly suggested pathogen-specific geographic and climatic variation, whereas differences according to breed purity and farming system were less consistent and should be interpreted with caution. The findings of this meta-analysis provide a long-term, global, empirical basis for veterinary epidemiological prevention and control, including the establishment of regional surveillance systems, the updating of control programs, and the rational use of antimicrobial agents. Furthermore, these findings provide key data-based support for genetic breeding for bovine mastitis resistance, including the need for molecular marker screening and regionally tailored breeding, as well as targeted nutritional regulation of mammary gland health. Overall, these findings may promote the interdisciplinary integration of animal genetic improvement, nutritional science, and epidemic prevention and control while offering new insights for global, healthy dairy farming and improved production performance. Despite evidence of potential publication bias or small-study effects for some pathogen categories, leave-one-out sensitivity analyses supported the robustness of the pooled global estimates. Future studies should further strengthen standardized regional surveillance and translate pathogen-specific distribution data into locally adapted breeding, nutritional, and management strategies. Supplementary Information Additional file 1: Table S1. PRISMA2020 checklist. Table S2. Literature search syntax for the pathogen spectrum of dairy cow mastitis. Table S3. Included literature and source information. Table S4. Inclusion and exclusion criteria for studies on the pathogen spectrum of dairy cow mastitis. Table S5. Risk of bias assessment based on STROBE criteria. Table S6. Bias assessment results for included studies based on STROBE criteria. Table S7. Quality assessment checklist. Table S8. Quality score and quality index score of individual contributing studies. Table S9. Data extraction variables from reviewed literature. Table S10. Bias in the inclusion of studies for multi-model validation. Table S11. Sensitivity analysis using alternative random-effects models for the five dominant pathogen categories. Table S12. Sensitivity analysis restricted to high-quality studies. Table S13. Research information and epidemiological situation of Enterococcus spp. causing bovine mastitis. Table S14. Research information and epidemiological situation of Pseudomonas spp. causing bovine mastitis. Table S15. Research information and epidemiological situation of Bacillus spp. causing bovine mastitis. Table S16. Research information and epidemiological situation of yeast causing bovine mastitis. Table S17. Research information and epidemiological situation of Corynebacterium spp. causing bovine mastitis. Table S18. Research information and epidemiological situation of Enterobacter spp. causing bovine mastitis. Table S19. Research information and epidemiological situation of Mycoplasma spp. causing bovine mastitis. Table S20. Research information and epidemiological situation of Trueperella spp. causing bovine mastitis. Table S21. Research information and epidemiological situation of Proteus spp. causing bovine mastitis. Table S22. Research information and epidemiological situation of Lactococcus spp. causing bovine mastitis. Fig. S1. Funnel plot for the meta-analysis of bovine mastitis caused by S. aureus. Fig. S2. Funnel plot for the meta-analysis of bovine mastitis caused by Streptococcus spp. Fig. S3. Funnel plot for the meta-analysis of bovine mastitis caused by E. coli. Fig. S4. Funnel plot for the meta-analysis of bovine mastitis caused by CNS. Fig. S5. Funnel plot for the meta-analysis of bovine mastitis caused by Klebsiella spp. Fig. S6. Period-specific hotspot countries for S. aureus infection in dairy cow mastitis. Fig. S7. Period-specific hotspot countries forStreptococcus spp. infection in dairy cow mastitis. Fig. S8. Period-specific hotspot countries for E. coli infection in dairy cow mastitis. Fig. S9. Period-specific hotspot countries for CNS infection in dairy cow mastitis. Fig. S10. Period-specific hotspot countries for Klebsiella spp. infection in dairy cow mastitis. Fig. S11. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S12. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S13. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S14. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S15. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S16. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S17. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S18. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S19. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S20. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S21. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S22. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S23. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S24. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S25. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S26. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S27. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S28. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S29. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S30. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S31. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S32. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S33. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S34. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S35. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S36. Spatiotemporal distribution of Enterococcus spp. causing dairy cow mastitis (2000–2025). Fig. S37. Spatiotemporal distribution of Pseudomonas spp. causing dairy cow mastitis (2000–2025). Fig. S38. Spatiotemporal distribution of yeast causing dairy cow mastitis (2000–2025). Fig. S39. Spatiotemporal distribution of Bacillus spp. causing dairy cow mastitis (2000–2025). Fig. S40. Spatiotemporal distribution of Corynebacterium spp. causing dairy cow mastitis (2000–2025). Fig. S41. Spatiotemporal distribution of Mycoplasma spp. causing dairy cow mastitis (2000–2025). Fig. S42. Spatiotemporal distribution of Trueperella spp. causing dairy cow mastitis (2000–2025). Fig. S43. Spatiotemporal distribution of Enterobacter spp. causing dairy cow mastitis (2000–2025). Fig. S44. Spatiotemporal distribution of Proteus spp. causing dairy cow mastitis (2000–2025). Fig. S45. Spatiotemporal distribution of Lactococcus spp. causing dairy cow mastitis (2000–2025). Abbreviations Bovine leukocyte antigen Confidence interval Coagulase-negative staphylococci Preferred Reporting Items for Systematic Reviews and Meta-Analyses Single nucleotide polymorphism Strengthening the Reporting of Observational Studies in Epidemiology Toll-like receptor 4 Authors’ contributions KY, WG and SF: Conceptualization, Methodology, and Software. KY, GH and ZP: Data curation, Writing-Original draft preparation. KY, GH, ZP, ML, YC, XL, JL and DH: Visualization, Investigation. WG and SF: Supervision. KY: Software and Validation. WG and SF: Writing – Reviewing and Editing. All authors read and approved the final manuscript. Funding This work was supported by the National Key Research and Development Program of China (Beijing, China; grant No. 2025YFF1000800); the Yunnan International Joint Laboratory for Exploration and Utilization of Aromatic Chinese Medicinal Materials in Agriculture (202503AP140024); and the National Natural Science Foundation of China (32372961, 32472984, and 32372959). Data availability The supplemental material, analytic code, original data, and findings of this study are available at https://doi.org/10.6084/m9.figshare.31120591. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Kai Yang and Guiqiu Hu contributed equally to this work. Contributor Information Wenjin Guo, Email: guowenjin@jlu.edu.cn. Shoupeng Fu, Email: fushoupeng@jlu.edu.cn. References Associated Data Supplementary Materials Additional file 1: Table S1. PRISMA2020 checklist. Table S2. Literature search syntax for the pathogen spectrum of dairy cow mastitis. Table S3. Included literature and source information. Table S4. Inclusion and exclusion criteria for studies on the pathogen spectrum of dairy cow mastitis. Table S5. Risk of bias assessment based on STROBE criteria. Table S6. Bias assessment results for included studies based on STROBE criteria. Table S7. Quality assessment checklist. Table S8. Quality score and quality index score of individual contributing studies. Table S9. Data extraction variables from reviewed literature. Table S10. Bias in the inclusion of studies for multi-model validation. Table S11. Sensitivity analysis using alternative random-effects models for the five dominant pathogen categories. Table S12. Sensitivity analysis restricted to high-quality studies. Table S13. Research information and epidemiological situation of Enterococcus spp. causing bovine mastitis. Table S14. Research information and epidemiological situation of Pseudomonas spp. causing bovine mastitis. Table S15. Research information and epidemiological situation of Bacillus spp. causing bovine mastitis. Table S16. Research information and epidemiological situation of yeast causing bovine mastitis. Table S17. Research information and epidemiological situation of Corynebacterium spp. causing bovine mastitis. Table S18. Research information and epidemiological situation of Enterobacter spp. causing bovine mastitis. Table S19. Research information and epidemiological situation of Mycoplasma spp. causing bovine mastitis. Table S20. Research information and epidemiological situation of Trueperella spp. causing bovine mastitis. Table S21. Research information and epidemiological situation of Proteus spp. causing bovine mastitis. Table S22. Research information and epidemiological situation of Lactococcus spp. causing bovine mastitis. Fig. S1. Funnel plot for the meta-analysis of bovine mastitis caused by S. aureus. Fig. S2. Funnel plot for the meta-analysis of bovine mastitis caused by Streptococcus spp. Fig. S3. Funnel plot for the meta-analysis of bovine mastitis caused by E. coli. Fig. S4. Funnel plot for the meta-analysis of bovine mastitis caused by CNS. Fig. S5. Funnel plot for the meta-analysis of bovine mastitis caused by Klebsiella spp. Fig. S6. Period-specific hotspot countries for S. aureus infection in dairy cow mastitis. Fig. S7. Period-specific hotspot countries forStreptococcus spp. infection in dairy cow mastitis. Fig. S8. Period-specific hotspot countries for E. coli infection in dairy cow mastitis. Fig. S9. Period-specific hotspot countries for CNS infection in dairy cow mastitis. Fig. S10. Period-specific hotspot countries for Klebsiella spp. infection in dairy cow mastitis. Fig. S11. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S12. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S13. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S14. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S15. Prevalence of S. aureus in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S16. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S17. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S18. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S19. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S20. Prevalence of Streptococcus spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S21. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S22. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S23. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S24. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S25. Prevalence of E. coli in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S26. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S27. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S28. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S29. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S30. Prevalence of CNS in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S31. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by country. Fig. S32. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by continent. Fig. S33. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by climate. Fig. S34. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by breed purity. Fig. S35. Prevalence of Klebsiella spp. in dairy cow mastitis: a meta-analysis and subgroup analysis by intensification. Fig. S36. Spatiotemporal distribution of Enterococcus spp. causing dairy cow mastitis (2000–2025). Fig. S37. Spatiotemporal distribution of Pseudomonas spp. causing dairy cow mastitis (2000–2025). Fig. S38. Spatiotemporal distribution of yeast causing dairy cow mastitis (2000–2025). Fig. S39. Spatiotemporal distribution of Bacillus spp. causing dairy cow mastitis (2000–2025). Fig. S40. Spatiotemporal distribution of Corynebacterium spp. causing dairy cow mastitis (2000–2025). Fig. S41. Spatiotemporal distribution of Mycoplasma spp. causing dairy cow mastitis (2000–2025). Fig. S42. Spatiotemporal distribution of Trueperella spp. causing dairy cow mastitis (2000–2025). Fig. S43. Spatiotemporal distribution of Enterobacter spp. causing dairy cow mastitis (2000–2025). Fig. S44. Spatiotemporal distribution of Proteus spp. causing dairy cow mastitis (2000–2025). Fig. S45. Spatiotemporal distribution of Lactococcus spp. causing dairy cow mastitis (2000–2025). Data Availability Statement The supplemental material, analytic code, original data, and findings of this study are available at https://doi.org/10.6084/m9.figshare.31120591.