Globally, organic agriculture has experienced substantial growth in recent decades, with growth rates of 20% per year in some countries (Rana and Paul, 2017; Willer and Lernoud, 2017). The United States (US) ranks third in worldwide organic industry production, accounting for approximately 2.2 million hectares of organic growing land and profits of nearly $43.3 billion per year (Willer and Lernoud, 2017). The rapid increase in consumer demand and consumption of organic produce has been attributed to the perceived benefits, both nutritionally and environmentally, of organic farming compared to conventional farming (Lotter, 2003).

Organic farming is based on the principles of health, ecology, fairness, and care, and focuses on sustainable practices (Luttikholt, 2007). Therefore, regulations prohibit organic farms from using synthetic fertilizers and pesticides, and organic producers must find other ways to supplement their fields. Application of biological soil amendments of animal origin (BSAAOs) and other nutrient and crop management techniques, such as the use of cover crops, crop rotation, and intercropping with nitrogen-fixing crops are frequently utilized by organic farmers (Rosen and Bierman, 2005).

As defined by the US Food and Drug Administration (FDA), BSAAOs are soil amendments, partially or wholly composed of animal waste and other byproducts, including, but not limited to, raw or untreated manure, animal carcasses, swine slurry, and poultry litter (FDA, 2015b). Use of these natural alternatives has numerous benefits for the soil quality, including increasing organic matter and nutrients in soil, water retention, drainage, and aeration (Rosen and Allan, 2007). However, many of the animal sources of such amendments (e.g., cattle, sheep, and poultry) are known sources of foodborne pathogens that are often linked to fresh produce outbreaks, such as Escherichia (E.) coli O157:H7, non-O157 Shiga toxin-producing E. coli (STEC), Salmonella spp., and Listeria (L.) monocytogenes (Whipps et al., 2008; Olaimat and Holley, 2012). Therefore, while the use of BSAAOs offers multiple benefits, it also risks introduction of foodborne illness pathogens into the field, creating potential for contamination of produce (Nightingale et al., 2004). Currently, certified organic producers follow US Department of Agriculture (USDA)-National Organic Program (NOP) standards for incorporation of raw manure to mitigate the contamination risk, and these differ between crops whose edible portions do and do not come into direct contact with the soil amended with raw manure (FDA, 2015b; USDA NOP, 2018). For crops not directly in contact with soil, e.g., tomatoes and peppers, a 90-day minimum interval between raw or incompletely treated manure incorporation and the crop harvest is required. In contrast, a 120-day minimum interval is required for crops that directly contact the soil, e.g., leafy greens and root vegetables (USDA NOP, 2018). However, these intervals are based on the production cycle and environmental factors, and do not take into account other on-farm factors that may contribute to pathogen survival and potential crop contamination. A multitude of factors, including pathogen, animal manure, and soil types, as well as environmental conditions, are known to influence the persistence of pathogens in the soil, and their potential to contaminate fresh produce (Harris et al., 2003; Hutchison et al., 2005; Erickson et al., 2014; Sharma et al., 2019; Limoges et al., 2022). There is currently minimal research to support the NOP interval guidelines, and the Produce Safety Rule (PSR) of the Food Safety Modernization Act (FSMA) has not specified its own interval standards for raw manure; instead, the FDA is waiting until more data has been collected to make a final ruling (FDA, 2015a, 2018).

The wide natural distribution of L. monocytogenes across environments and hosts provides extensive opportunities for transmission to foods, including fresh produce specialty crops that are consumed without cooking. L. monocytogenes has been isolated in non-agricultural and agricultural environments (Linke et al., 2014; Harrand et al., 2020; Liao et al., 2021). For example, manure contaminated with L. monocytogenes used to fertilize a field can introduce the pathogen into the soil (Jiang et al., 2004). Additionally, studies have shown that L. monocytogenes may survive at detectable levels in soil for 128 days after manure application, allowing for potential contamination of produce by soil cultivation, splash events or contaminated water (Weller et al., 2016; Harrand et al., 2020). Foodborne illness outbreaks caused by L. monocytogenes, while relatively uncommon, can result in high morbidity and mortality, especially in pregnant, immunocompromised, and elderly individuals, with case fatality rates reaching approximately 20% (CDC, 2013; Pohl et al., 2019). While many listeriosis outbreaks are linked to contaminated animal products (e.g., deli meats, soft cheeses), several outbreaks have involved fresh produce as the vehicle of transmission in the US over the past decade, including cantaloupes in 2011, bean sprouts, caramel apples, nectarines, and peaches in 2014, fresh packaged salads in 2016 and 2021, and enoki mushrooms in 2020 and 2022 (CDC, 2013, 2023; Garner and Kathariou, 2016; Townsend et al., 2021).

Studies have shown that environmental factors have a positive association with L. monocytogenes detection and isolation (e.g., soil, water, vegetation) (Ivanek et al., 2009; Harrand et al., 2020). Additionally, adjacent land use for livestock pastures was associated with the prevalence of L. monocytogenes in agricultural fields (Harrand et al., 2020). Manure application, soil cultivation, and wildlife intrusion have also been shown to increase the risk of L. monocytogenes contamination of produce fields (Strawn et al., 2013b). However, many studies reporting on such risks were conducted on conventional farms and/or in controlled trials, often missing organic-specific variables, such as management practices and environmental conditions of organic production systems.

Longitudinal, multi-regional studies are necessary to establish the complex time-intervals between untreated manure application and harvest, and to identify specific risk factors associated with persistence of L. monocytogenes on organic farms using untreated manure-based soil amendments. In order to provide a more complete understanding and approach to risk mitigation and best practices for the use of untreated manure (i.e., manure that has not been processed to completion in a scientifically valid manner to reduce the presence of microorganisms of public health significance) in organic production, this study focused on a broad range of variables that may impact contamination risk in various regions, and was informed by multiple disciplines. Specifically, the objective of this study was to assess farm-level risk factors associated with L. monocytogenes prevalence in soils amended with untreated manure on USDA-NOP certified organic farms in four regions of the US.

Of 2,150 soil samples, 101 (4.7%) were positive for L. monocytogenes. None of the fresh produce or irrigation water samples were positive for L. monocytogenes.

This longitudinal study is one of the first to simultaneously assess multiple risk factors, including farm management practices and demographics, environmental factors (i.e., soil and microbiological characteristics), and weather factors, associated with the presence of L. monocytogenes in soil amended with untreated manure on certified organic farms in four different regions of the US. Previous studies have found that farm management, landscape, meteorological, and spatio-temporal factors significantly impact the odds of L. monocytogenes contamination of produce fields (Strawn et al., 2013a,b; Weller et al., 2015). However, these studies were restricted to a single state/region of the US, used a limited number of samples collected in the same field/plot, and did not specifically focus on certified-organic farms. Additionally, this study provides scientific insight into the complex nature of L. monocytogenes persistence in the soil, and subsequent contamination of the farm environment, demonstrating the need for food safety guidelines to be data-driven and to account for more than a single metric of time interval between manure application and harvesting of fresh produce.

The odds of L. monocytogenes presence in the soil were significantly lower in 2018 than in 2017, which is similar to the patterns reported for non-O157 STEC on the studied farms (Pires et al., 2023). The majority of farm management practices remained the same during the 2 years, thus, the difference may be attributed to changes that could not be controlled, such as weather conditions, soil characteristics, and/or other environmental factors. Notably, 2017 was the second wettest year on record in CA, where the majority of the study farms were located, with some areas receiving 300% of the average precipitation, ending the state’s 5-year drought (California Department of Water Resources, 2017), whereas another statewide drought was recorded in 2018 (California Department of Water Resources, 2018). Runoff, especially from livestock farms, has previously been shown to increase the risk of L. monocytogenes contamination, and the historic amount of runoff in CA in 2017 may, therefore, have contributed to the increased presence of L. monocytogenes in the soil (Strawn et al., 2013a). Moreover, other studies have demonstrated a positive relationship between adjacent land use, specifically for livestock farming, and L. monocytogenes contamination (Sauders et al., 2006; Lyautey et al., 2007). In contrast, the other states involved in the project had normal or increased precipitation in 2018 when compared to 2017. These differences show the critical need for multi-year studies given the weather conditions may drastically change across the years.

In both models 1 and 2, sampling day was significantly associated with the presence of L. monocytogenes in the soil. While all sampling days had decreased odds of L. monocytogenes contamination when compared to d0B, d180 had the lowest. Previous studies have investigated the prevalence of L. monocytogenes in the soil over time (Hutchison et al., 2004; Locatelli et al., 2013). While Hutchison et al. (2004) found the maximum isolation time for L. monocytogenes was 120 days, other studies did not detect the pathogen beyond 2 weeks (Hutchison et al., 2004; Locatelli et al., 2013). Differences between these isolation times were attributed to environmental factors (e.g., climate, temperature, spatial coordinates) and soil characteristics (e.g., pH, soil chemistry, soil texture) (Locatelli et al., 2013); likewise we identified weather and soil factors associated with the presence of this foodborne pathogen. Certified organic farms follow the 90–120 days USDA-NOP standards time-interval between the application of raw manure and crop harvest, depending if the edible part is in contact or not with the soil (USDA NOP, 2018). The time-interval as a preventive measure to avoid contamination of the soil and produce is based on the assumption that an appropriate amount of time has passed to foodborne pathogen die-off (Phan-Thien et al., 2020). However, our findings suggest that the NOP 90- and 120-day wait times may be sufficient, and there was a significant decrease compared to 30 days post manure application; this time interval may be an impairment to the use of untreated manure in regions with multiple growing seasons, such as California. Moreover, this study shows that a range of on-farm and environmental factors are associated with the presence of foodborne pathogens in soil, indicating that those risk factors should be taken into account in pre-harvest food safety plans.

Regarding the microbiological characteristics of the soil, previous studies have shown that generic E. coli can serve as an indicator for foodborne pathogens in soil, including L. monocytogenes (Falardeau et al., 2017). Similarly, we observed increased concentrations of generic E. coli in the soil were associated with increased odds of L. monocytogenes presence. By contrast, no positive association was observed with generic E. coli concentrations in water, which confirms that it may not be a suitable predictor for L. monocytogenes in water, given the complex interactions between weather, and water microbial and physicochemical characteristics (Weller et al., 2020). The presence of other pathogens (i.e., non-O157 STEC, E. coli O157:H7, and Salmonella spp.) and of L. monocytogenes in manure applied to crop fields was associated with increased odds of L. monocytogenes in the soil. While some studies have found that L. monocytogenes survival is increased in the absence of normal competitors, they did not include other foodborne pathogens (Botzler et al., 1960; Dowe et al., 1997; McLaughlin et al., 2011). It is possible that characteristics of the manure, such as its percentage of organic matter, rather than microbial activity, influenced the presence of L. monocytogenes in the soil. Previous studies have noted potential L. monocytogenes transmission pathways from infected livestock feces/manure to soil (Nightingale et al., 2004). However, further analyses are needed to determine whether the strains of L. monocytogenes isolated from the soil in this study were the same as those present in the untreated manure applied to the field. Other studies have detailed the survival of Listeria spp. in the soil and established how the bacteria may be transferred to produce through contact with soil or splashing events caused by rain or irrigation, thereby highlighting the food safety risks associated with contaminated soil (Girardin et al., 2005; Oliveira et al., 2011). Additionally, L. monocytogenes is known to interact with the roots of leafy produce, and may internalize into mature plants as well as seedlings, but relatively low concentrations of L. monocytogenes have been isolated from contaminated fresh produce (Standing et al., 2013; Smith et al., 2018). Further studies are needed to determine which levels of L. monocytogenes in the raw manure may lead to the contamination of soil and subsequently to that of produce in fields amended with raw manure. Aging or composting manure before application to produce fields is a promising way to significantly decrease pathogen activity in manure, and thus decrease the risk of soil, and subsequent produce contamination (Gurtler et al., 2018).

Surprisingly, none of the variables related to the presence of domestic animals or wildlife were significantly associated with the presence L. monocytogenes in the soil, when they are known sources of foodborne pathogen contamination in produce fields (Hoelzer et al., 2011; Strawn et al., 2013a). However, it is important to note that domestic animal and wildlife presence were self-reported by farmers, and therefore may not have been accurately evaluated. The presence of fecal matter on the field may provide a more valid assessment of the presence of animals. However, while it was positively associated with the presence of non-O157 STEC in the soil, it was not associated with L. monocytogenes positivity (Pires et al., 2023).

Of the numerous farm management and demographic variables assessed in this study, only season of manure application, number of year-round staff, and water source, were significantly associated with L. monocytogenes presence in the multivariable models. Manure application in winter and summer was associated with higher and lower odds of L. monocytogenes positivity, respectively, when compared to spring. Unlike many other pathogens, L. monocytogenes has been found to survive, and even be at its highest prevalence, in winter months and conditions (Botzler et al., 1960; Guerini et al., 2007; Strawn et al., 2013a; Cooley et al., 2014). Additionally, previous studies reported an increase in L. monocytogenes survival in soil with decreased competition with native soil microbiota (Botzler et al., 1960; Dowe et al., 1997; McLaughlin et al., 2011). In winter months, other pathogens, such as Salmonella spp., persist at much lower rates than during warmer months, thus decreasing microbial competition for L. monocytogenes (Cooley et al., 2014). The increased survival of L. monocytogenes in winter months may contribute to the increased risk of L. monocytogenes transmission from untreated manure to the soil of the fields to which it is applied. Due to state and local regulations, or to follow the NOP 90- or 120-day time interval, the majority of farms apply manure to the fields in the fall, during which the odds of L. monocytogenes presence was lower (i.e., not significantly different from spring) (Pires et al., 2018).

We found that farms with five or more year-round staff have lower odds of L. monocytogenes presence in soil than smaller farms with four or fewer staff. A previous study in the southwest US found that farm hygiene was a protective factor of generic E. coli contamination in produce; however, this study did not explore soil contamination, or the effect the size that the farming operation had on contamination (Park et al., 2014). Larger farms with increased numbers of employees are more likely to participate in third party food safety certification programs or audits, such as GLOBALG.A.P (Schmit et al., 2020). Thus, larger farming operations are more likely to have a strong food safety culture and training programs that allow better implementation of measures that decrease the risk of microbial contamination of the fields and produce. Additionally, Ramos et al. (2019) reported that many small farmers find on-farm food safety plans and training intimidating without the assistance of third parties. This lack of employee training on smaller farms may result in inadequate application of Good Agricultural Practices, thus increasing the risk of L. monocytogenes contamination in produce fields. Additional outreach and educational support for these smaller farms may help decrease the risk of pathogen contamination of soil and fresh produce.

Water is a known reservoir of, and can serve as a transmission pathway for foodborne pathogens to soil and produce (Beuchat, 2006). In univariable analysis, only surface water significantly differed from well water, and it had a surprising negative association with L. monocytogenes presence in the soil. However, no positive surface water samples could be included in multivariable analysis due to the decrease in sample size. In multivariable analysis, we found that the use of mixed sources of irrigation water was positively associated with L. monocytogenes presence in the soil, whereas the use of municipal sources had a protective effect, although it was borderline significant. This could be due to differences in environmental contamination among water sources, as engineered water sources, such as municipal or well water, are generally more protected than surface and mixed irrigation water sources. Strawn et al. (2013a) indeed found that surface water was significantly more likely to be contaminated with L. monocytogenes than water from engineered sources. In order to reduce the contamination risks and food safety hazards from agricultural water, FSMA-PSR recently established requirements for microbial standards; however, the pre-harvest agricultural water rule is still under review (FDA, 2015a).

Interestingly, we found that the presence of L. monocytogenes in the soil was negatively associated with the 7-day temperature range, but none of the associations with other weather variables were significant in the multivariable models. This contrasts with previous studies, which found that higher precipitation levels increased the survival of L. monocytogenes (Park et al., 2014; Pang et al., 2017). Further investigation into the short- and long-term effects of precipitation on soil contamination by L. monocytogenes may offer better insights into the risk posed by this weather event. L. monocytogenes is known to survive the large temperature fluctuations associated with food production and processing (Liu, 2008; Smith et al., 2018), which may explain the negative association we observed with the 7-day temperature range. However, the absence of association with other temperature variables conflicts with previous findings, as L. monocytogenes survival is generally increased at cooler temperatures (McLaughlin et al., 2011; Strawn et al., 2013a).

Soil moisture increases with decreased temperature and increased precipitation, which favors L. monocytogenes survival (Botzler et al., 1960; Weis and Seeliger, 1975; Strawn et al., 2013a). Similarly, four of the six HOBO soil moisture variables were significantly negatively correlated with L. monocytogenes presence in the soil, with long-term (i.e., 7 days) variables showing slightly stronger effects than short-term (i.e., 0 days) variables. However, this contrasts with this study’s findings for soil moisture at the sample-level, which showed a significant negative association with L. monocytogenes presence.

Regarding soil composition, we found that an increase in the percentage of organic matter in the soil was associated with an increased odds of L. monocytogenes presence, which is similar to previous studies (Jamieson et al., 2002; Vogel et al., 2010; Strawn et al., 2013a). This increase in pathogen survival has been attributed to the effects of organic matter on soil moisture retention (Jamieson et al., 2002; Williams et al., 2015). However, we observed a negative association between soil moisture at the sample-level and L. monocytogenes presence. It is important to note that the addition of manure to the field may have a confounding effect as it increases the organic matter content of the soil, and may also introduce L. monocytogenes into the soil. Calcium concentration in the soil was positively associated with the presence of L. monocytogenes. This observation is similar to previous findings that identified calcium as one of the main soil characteristics that impacted L. monocytogenes survival in soil (Locatelli et al., 2013). Nitrate and potassium concentrations in the soil were positively and negatively associated with the presence of L. monocytogenes, respectively. A recent study reported that an increase in macronutrients in the soil, namely nitrogen, potassium, and phosphorus, was associated with a decrease in L. monocytogenes survival (Devarajan et al., 2021).

Of all the soil characteristics included in the final multivariable model, only calcium had a significant effect in the univariable analysis, indicating the presence of confounding factors among the variables. The complex nature of soil chemistry, and the intricate relationships among nutrients and soil characteristics (e.g., pH, organic matter) render it difficult to identify which variables are interacting and how they affect L. monocytogenes survival. Furthermore, research into the effects of soil chemical characteristics on L. monocytogenes presence in the soil is scant, with many studies identifying numerous interactions among the soil characteristics investigated, and others focusing only on soil textures (Dowe et al., 1997; Nicholson et al., 2005; Locatelli et al., 2013). Further studies are needed to identify the individual and combined effects of soil characteristics as well as the role the microbiome plays on the survival of pathogens in the soil.

On the one hand, one of the advantages of on-farm observational studies with repeated sampling (i.e., longitudinal studies), is that they allow to investigate the dynamics of foodborne pathogens in natural settings (as opposed to inoculation studies), with contamination loads in environment matrixes (e.g., manure and soil) as close as possible to those on commercial settings, and to account for time-variable factors. On the other hand, there are limitations inherent to this type of study design. First, because there are no interventions across the farms (besides manure application), some of the key factors affecting the survival of pathogens may vary (e.g., soil type, weather, irrigation); however, those were controlled for in statistical analysis. Second, the loss due to follow-up may contribute to an unbalanced sampling across the 2 years. In addition, other limitations are inconsistent sample collection for soil characteristic testing (due to scheduling limits, weather prohibiting sampling, unpredictable farming schedules, etc.), as well as missing weather data, which restricted some of the statistical analyses. Inconsistent sampling and missing data also contributed to an unbalanced data set for some variables. Moreover, data groups were collected with different temporal resolutions (sampling day, several times a year) and spatial resolutions (farm versus plot); this was partially accounted for by using random effects. These are limitations of observational studies conducted on commercial farms as opposed to controlled experimental field trials. In addition, we could not assess the genetic relatedness of the L. monocytogenes isolates detected in manure and respective amended soils. However, as an observational study, the goal was to assess the multitude of risk factors associated with the presence and persistence of this foodborne pathogen in USDA-NOP certified organic farms using untreated manure as a soil amendment. Additionally, detection of L. monocytogenes was limited by the sensitivity of the testing protocols used in this study. It is finally important to note that this study allowed us to report only observed associations and not to infer causal relationships.

This is the first study to examine L. monocytogenes contamination risk factors associated with untreated-manure–amended soil on organic farms in four regions of the US. The findings of this study highlight that factors beyond time intervals are associated with the presence of L. monocytogenes in soil. While sampling time was significantly associated with L. monocytogenes contamination, multiple environmental (weather, soil characteristics), management (manure application season, number of employees), and microbiological (generic E. coli, other foodborne pathogens) factors were significantly associated with the presence of L. monocytogenes in the soil. This emphasizes the need to further examine risk factors beyond time intervals, and incorporate those variables into regulations for the use of untreated manure on fresh produce fields.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent from the participants was not required to participate in this study in accordance with the national legislation and the institutional requirements.

AFAP, MJ-R, PM, PP, MH, JS, and BM-L conceived the project, designed the study, and wrote the grant. AFAP, PM, FH, PP, and MH coordinated the efforts between the regional teams. AFAP oversaw the day-to-day aspects of the project. AFAP, MJ-R, and PM oversaw the laboratory work. TR, AK, CC-K, VH, and PA carried out the field work and sampling collection. TR, VH, and PA carried out the laboratory work. AFAP, JB, and BM-L developed the data analysis plan, which JB implemented. CC-K, JF, JB, and VH wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the final manuscript.

This work was funded by the Organic Agriculture Research and Extension Initiative (OREI), National Institute of Food and Agriculture, U.S. Department of Agriculture, under award number 2016-51300-25724.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.