Table 1.
The relevant studies used in the review.
| Reference | Study Period | Location | Sample Size | Analytical Method | Association Found |
|---|---|---|---|---|---|
| [5] | 1989–1996 | England and Wales (Europe) | Over 52,000 (exclude 1999 cases who had recent foreign travel) | Ordinary least-squares regression | A positive association between monthly cryptosporidiosis rates and temperature in the previous month from August to November. A negative association between monthly cryptosporidiosis rates and precipitation in the previous month from August to November. A positive association between monthly cryptosporidiosis rates and maximum river flow in the current month between April and July. A positive association between monthly cryptosporidiosis rates and maximum river flow in the current month between August and November, only when temperature and precipitation in the previous month were included in the model. |
| [6] | 1996–2004 | Australia | NA | Time series Poisson regression and seasonal auto-regression integrated moving average (SARIMA) models | A positive association between maximum temperature at lags of 1 to 3 months and Cryptosporidium. A positive association between relative humidity at a lag of 1 month and Cryptosporidium. |
| [7] | 2005–2008 | India (Asia) | 2579 | Mann–Whitney U test and Fisher’s exact test | A positive association between temperature and Cryptosporidium. A negative association between humidity and Cryptosporidium. A higher rate of cryptosporidiosis positivity during hotter and drier weather especially in Delhi. |
| [8] | 2001 | Australia | NA | Three-stage spatiotemporal classification and regression tree (CART) models | A positive association between temperature and the incidence of cryptosporidiosis for the database without zero incidences. No association between temperature and the incidence of cryptosporidiosis for the full database in any season. A positive association between rainfall and Cryptosporidium. |
| [9] | 1996–2004 | Australia | NA | Time series zero-inflated Poisson (ZIP) and classification and regression tree (CART) models Cross-correlation function |
A negative association between rainfall at lags of 0–1 week and cryptosporidiosis incidence. A negative association between wind speed at a lag of 4 weeks and cryptosporidiosis incidence. A positive association between maximum temperature at lags of 0–8 weeks and cryptosporidiosis incidence. The incidence of cryptosporidiosis might increase at temperatures above 31 °C and relative humidity below 63%, otherwise the effect on cryptosporidiosis was insignificant |
| [10] | 1997–2008 | New Zealand (Oceania) | 8092 cases of cryptosporidiosis and 10,424 cases of giardiasis | Seasonal Auto Regressive Integrated Moving Average (SARIMA) models | No climatic factors were significantly associated with giardiasis. Cryptosporidiosis was positively associated with the average temperature of the previous month. |
| [11] | Human: 2009–2015, Bovine: 2008–2014 |
Canada (North America) | 214 bovine and 87 human cases |
Logistic regression models and a case-crossover approach | A positive association between the maximum ambient air temperature at 0, 9, or 14 days and human cryptosporidiosis. No association between rainfall and the incidence of human cryptosporidiosis in this study. A negative association between average water flow and the incidence of human cryptosporidiosis. |
| [12] | 2004–2016 (cryptosporidiosis) | Six Northern/Arctic countries or country parts | NA | Spearman’s correlation coefficient and stepwise regression | A positive association between the incidence of cryptosporidiosis and autumn temperature. A positive association between the incidence of cryptosporidiosis and mean temperature of the wettest quarter. A positive association between the incidence of cryptosporidiosis and annual maximum monthly precipitation. |
| [13] | 2001 to 2009 | Australia | NA | A negative binomial regression model | A positive association between the mean maximum monthly temperature lagged by 2 months and cryptosporidiosis notifications in metropolitan areas. A positive association between the mean minimum monthly temperature and cryptosporidiosis notifications in metropolitan areas. A negative relationship between the mean maximum monthly temperature lagged by 3 months and cryptosporidiosis notifications in rural areas. A negative relationship between the mean minimum monthly temperature lagged by 3 months and cryptosporidiosis notifications in rural areas. No association was found with rainfall for any area examined. |
| [14] | 2001–2003, 2008–2010 | Tanzania (Africa) | 406 | Multivariate logistic regression analysis |
A positive association between rainfall and Cryptosporidium. A positive association between maximum temperature and Cryptosporidium. |
| [15] | Water samples: 2003 | Australia | NA | Liner regression | Temperature is a critical parameter in the survival and infectivity of oocysts shed into the environment. Higher temperatures decrease autonomous survival and infectivity. |
| [16] | 2004–2008 | Canada (North America) | 1171 | CART (Classification and Regression Tree) and binary logistical regression techniques Spearman’s rank correlation |
Cryptosporidium and Giardia oocyst and cyst densities were positively associated with surface water discharge, and negatively associated with air/water temperature during spring–summer–fall. Some of the highest Cryptosporidium oocyst densities were found to be associated with low discharge conditions on small stream orders, suggesting wildlife as a contributing fecal source. |
| [17] | NA | America (North America) | NA | Multivariable regression analysis | A positive association between temperature and die-off of Cryptosporidium. A positive association between temperature and die-off of Giardia. The direct correlation between higher die-off rates of parasites (oo)cysts and higher porosity of surfaces. |
| [18] | 1997 to 2006 | New Zealand (Oceania) | NA | Negative binomial regression | A negative association between average annual temperatures and cryptosporidiosis in both the source model and distribution mode. A weakly positive association between average annual temperatures and giardiasis that only reached borderline significance in the distribution model. A positive association between rainfall and giardiasis/cryptosporidiosis (both models). Urban/rural status was a strong predictor of cryptosporidiosis in both the univariate and multivariate regression analyses. |
| [19] | NA | Data from 19 countries in Central and South America, Sub-Saharan Africa and South and Southeast Asia |
NA | An IPD-MA framework, modified Poisson models with robust variance estimation, generalized linear models (GLMs) |
No association between relative humidity and Cryptosporidium. A slight inverse association between relative humidity and Giardia. A positive association between soil moisture and Cryptosporidium. A slight inverse association between solar radiation and Cryptosporidium or Giardia. The adjusted association between temperature and Cryptosporidium was a skewed, inverted U-shaped relationship, while the association with Giardia took on a more sinusoidal shape. |
| [20] | 2016 | Canada (North America) | 55 water samples | Exact logistic regressions | Cumulative precipitation, water level, and turbidity were not associated with the presence/absence of parasites. Low water/air temperature increased the possibility of the presence/absence of parasites. |
| [21] | 2012–2013 | Central California (North America) | NA | Logistic regression, Poisson regression |
Cryptosporidium oocyst concentrations were negatively associated with 30-day mean wind speed and cumulative precipitation. Giardia spp. cyst concentrations were positively associated with turbidity and the pH of water, and negatively associated with 24 h mean air temperature. |
| [22] | 1997–2009 | Canada (North America) | 7422 | Poisson regression and a distributed lag nonlinear regression model (DLNM) | No association between temperature and cryptosporidiosis cases was reported. No association between temperature and giardiasis cases was reported. A positive association between extreme precipitation with a lag of 4–6 weeks and cryptosporidiosis & giardiasis. |
| [23] | 1996 | America (North America) | NA | Spearman rank correlations | A positive association between extreme weather events and Giardia cysts. A significant correlation of Giardia and Cryptosporidium concentrations with turbidity levels. Cryptosporidium oocyst concentrations were associated with 12 other parameters, 9 of which also were associated with Giardia. Of the associated parameters, the ones that had the highest correlations with parasite concentrations were coliphage, DO, air, and water temperature (Giardia only), total and fecal coliforms, E. coli, alkalinity, hardness, daily average pH, river flow (Cryptosporidium only), and turbidity (both parasites). |
| [24] | 2001–2018 | Australia | NA | A Bayesian spatio-temporal analysis | A positive association between heavy/extreme rainfall and Cryptosporidium. |
| [25] | 2000–2002 | America (North America) | 193 | Kruskal–Wallis non-parametric test, a time series analysis (PROC ARIMA) | A positive association between rainfall and Cryptosporidium. Although mean Giardia cyst peaks occasionally coincided with rainfall peaks, the lowest monthly rainfall occurred in March of the first year when cyst levels were lowest. The sample cross-correlation functions revealed a peak in positive correlation with rainfall at lag zero, with negative correlation near the 2 month time lag, forward and backward. |
| [26] | 2005–2007 | Paris in France (Europe) | 162 | The nonparametric Spearman’s rho, the nonparametric Wilcoxon test | A weak and positive association between rainfall and Giardia. No association between rainfall and Cryptosporidium. |
| [27] | 2002–2004 | California (North America) | 350 | A negative binomial regression model | Vegetation buffer zones significantly reduced Giardia cysts in storm runoff. A negative association between cumulative precipitation and the concentration of Giardia. |
| [28] | 2012–2013 | India (Asia) | NA | Separate multivariable models of protozoa contamination | Rainfall sometimes increased the Giardia concentration and sometimes diluted it. |
| [29] | 2006–2013 | Canada (North America) | 403 cases | A Poisson multivariable regression model | An inverse association between Giardia and water level lagged by 1 month. An inverse association between within-stratum highest precipitation levels occurring 4 weeks prior to human cases and giardiasis occurrence in the truncated time series. |
| [30] | 1997–2005 | New Zealand (Oceania) | NA | Time-series studies, ordinary least-squares regression | In the summer and autumn, the cryptosporidiosis rate was positively associated with temperature in the current and previous month. No association between rainfall and cryptosporidiosis. |
| [31] | NA | America (North America) | NA | Linear regression (t- and F-tests) | The use of artificial ultraviolet (UV) light for oocyst disinfection. |
| [32] | NA | Hawaii (North America) | NA | Linear regression | A 90% reduction was identified in the viability of Cryptosporidium in marine waters after a 3 day exposure to solar inactivation. Giardia survived longer in canal water than in dark seawater, indicating that salinity has a greater effect on its inactivation than water quality. |
| [33] | 2005–2006 | Australia | NA | Linear regression | Solar UV can rapidly inactivate C. parvum in environmental waters. |
| [34] | NA | NA | 1 × 107 purified C. parvum oocysts or 5 × 105 of G. muris cysts |
A test of comparison of proportion, a pair-wise multiple comparison procedure and one-way ANOVA | Results showed that cysts of G. muris and oocysts of C. parvum are rendered completely noninfective after batch SODIS exposures of 4 and 10 h, respectively, and is also likely to be effective against waterborne cysts of Giardia lamblia. |
| [35] | 2007 | Goromonzi in Zimbabwe (Africa) | NA | NA: Control experiment (the average results with their standard deviations) | There is a synergistic effect between the heat and the UV light produced by the sun on the inactivation of Giardia. |
| [36] | NA | Soil samples: Sydney (Australia) | NA | Fluorescence in situ hybridization (FISH) and the Student–Newman–Keuls Test | Soil type emerged as a significantly influential factor for Cryptosporidium inactivation. |
| [37] | NA | Water samples: The State of Israel (Asia) | NA | NA: Control experiment | The die-off of C. parvum in saturated and dry loamy soil was monitored over time using immunofluorescence assay (IFA) and PCR to estimate oocysts viability and by cell culture to estimate oocysts infectivity. |
| [38] | NA | Southeastern New York State (North America) | 782 | Logistic regression | A negative association between the pH of the soil and the likelihood of detecting Cryptosporidium oocysts. Vegetation at the sampling site was significantly associated with the risk of detecting Giardia in the soil, and areas that were brush or bare soil were less likely to test positive for Giardia than land that had managed grass. A positive association between soil moisture and Giardia cysts. |
| [39] | NA | Western Australia | NA | NA: sludge storage trials and small scale soil amendment trials | A positive association between soil moisture and Giardia cysts. |
| [40] | NA | NA | NA | NA: control experiment | Cryptosporidium oocysts thrive better in soils with alkaline pH than in those with acidic pH. |
| [41] | NA | Australia | NA | Analysis of variance and analysis of covariance with the SAS generalized linear model procedure simulation experiment | Land slope was an important factor affecting the concentration of oocysts in runoff. |
| [42] | 2006–2007 | New York (North America) | NA | Spearman rank statistical test | A positive relationship between the occurrence of Cryptosporidium oocysts and extreme weather events. A positive relationship between the occurrence of Giardia cysts and extreme weather events. |
| [43] | 15 months | Germany (Europe) | NA | NA: control experiment | A risk of water-related infections through water-bound activities exists, especially after rain events and in times when thunderstorms are common. |
| [44] | 2002–2004 | America (North America) | NA | A negative binomial regression model | Vegetation cover can effectively reduce the amount of Cryptosporidium transfer from terrestrial manure to water. |