If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Bacterial spores, which are found in raw milk, can survive harsh processing conditions encountered in dairy manufacturing, including pasteurization and drying. Low-spore raw milk is desirable for dairy industry stakeholders, especially those who want to extend the shelf life of their product, expand their distribution channels, or reduce product spoilage. A recent previous study showed that an on-farm intervention that included washing towels with chlorine bleach and drying them completely, as well as training milking parlor employees to focus on teat end cleaning, significantly reduced spore levels in bulk tank raw milk. As a follow up to that previous study, here we calculate the costs associated with that previously described intervention as ranging from $9.49 to $13.35 per cow per year, depending on farm size. A Monte Carlo model was used to predict the shelf life of high temperature, short time fluid milk processed from raw milk before and after this low-cost intervention was applied, based on experimental data collected in a previous study. The model predicted that 18.24% of half-gallon containers of fluid milk processed from raw milk receiving no spore intervention would exceed the pasteurized milk ordinance limit of 20,000 cfu/mL by 17 d after pasteurization, while only 16.99% of containers processed from raw milk receiving the spore intervention would reach this level 17 d after pasteurization (a reduction of 1.25 percentage points and a 6.85% reduction). Finally, a survey of consumer milk use was conducted to determine how many consumers regularly consume fluid milk near or past the date printed on the package (i.e., code date), which revealed that over 50% of fluid milk consumers surveyed continue to consume fluid milk after this date, indicating that a considerable proportion of consumers are exposed to fluid milk that is likely to have high levels spore-forming bacterial growth and possibly associated quality defects (e.g., flavor or odor defects). This further highlights the importance of reducing spore levels in raw milk to extend pasteurized fluid milk shelf life and thereby reducing the risk of adverse consumer experiences. Processors who are interested in extending fluid milk shelf life by controlling the levels of spores in the raw milk supply should consider incentivizing low-spore raw milk through premium payments to producers.
Bacterial spores are found ubiquitously in the dairy farm environment and can survive processing hurdles that are commonly used in the dairy industry, such as pasteurization and drying (
in: De Vos P. Garrity G.M. Jones D. Krieg N. Ludwig W. Rainey F.A. Schleifer K. Whitman W.B. Bergey’s Manual of Systematic Bacteriology. 3rd ed. Springer US,
2009: 21-128
). The primary biological factor that limits HTST pasteurized fluid milk shelf life in the absence of product recontamination after pasteurization (i.e., postpasteurization contamination with gram-negative bacteria) is spore-forming bacteria capable of growing at refrigeration temperatures, otherwise known as psychrotolerant spore-forming bacteria (
). Psychrotolerant spore-forming bacteria have been shown to be responsible for approximately 50% of HTST fluid milk reaching the pasteurized milk ordinance (PMO) bacterial limit of 20,000 cfu/mL during refrigerated shelf life (
Pseudomonas fluorescens group bacterial strains are responsible for repeat and sporadic postpasteurization contamination and reduced fluid milk shelf life.
). For example, the psychrotolerant spore-forming bacteria Bacillus weihenstephanensis is responsible for sweet curdling, a coagulation defect in fluid milk (
). Shelf-life limitations caused by spore-forming bacteria restrict processors' abilities to implement distribution channel efficiencies (e.g., reducing the number of visits to an individual retailer) and enter new markets (e.g., eCommerce channels), which increases the overall risk that consumers will have a negative product experience. Ultimately the intrinsic quality of a food product (e.g., flavor, odor, and texture) influences consumer acceptance and repeat purchase of that food product (
report that consumers who experience reduced quality food products may stop purchasing that brand of product or even stop purchasing that category of products altogether (
). For the contemporary dairy industry, this could mean that consumers who have a negative experience with fluid milk of poor quality may turn to milk alternatives (e.g., almond-based beverages).
Current control strategies for reducing dairy product spoilage by spore-forming bacteria focus primarily on processing interventions. For example, centrifugation (i.e., bacterial clarification) and microfiltration are both mechanical methods used in the dairy industry to remove spores from milk before pasteurization (
). Additionally, studies have shown that reducing HTST pasteurization temperatures results in reduced outgrowth of spore-forming bacteria in fluid milk (
). In addition to processing strategies to reduce spore levels and outgrowth, a growing body of evidence suggests that reducing the incoming level of spores in raw milk by controlling on-farm management practices is a viable option for obtaining low-spore raw milk and subsequently reducing or delaying product spoilage. For example, recent studies have found that removing udder hair and using inorganic bedding, such as sand, are associated with lower spore counts in raw milk (
examined an on-farm spore intervention strategy that included washing towels used for milking preparation with bleach, drying towels, and giving a onetime training to milking parlor employees on the importance of teat end cleanliness, which was found to lower the spore count in bulk tank raw milk. Using this intervention, a 37% and 40% decrease in bulk tank raw milk mesophilic spores and thermophilic spores was observed, respectively, compared with spore levels in raw milk before the intervention was applied. Here, in the current study, we assessed the costs of the interventions described and applied in
, as well as their positive impacts on reducing fluid milk spoilage and enhanced customer satisfaction. While it has been demonstrated that the intervention strategy described in
is effective at reducing bulk tank raw milk spore levels, the cost of the intervention may be a barrier to implementation at the production level. Therefore, the objective of this study was to determine the cost to implement an on-farm spore intervention strategy previously shown to significantly reduce spores in bulk tank raw milk, as well as to determine the impact of the intervention strategy on the predicted levels of spore-forming bacteria throughout fluid milk shelf life using a Monte Carlo simulation. Finally, we surveyed consumers about in-home consumption habits to determine what proportion of consumers continue to consume HTST fluid milk past the date printed on the package (i.e., code date), potentially exposing them to reduced quality fluid milk. We hypothesized that the previously defined effective, easy-to-implement on-farm strategy (
) will be cost effective for producers to implement, provide a considerable improvement in predicted fluid milk shelf life, and reduce the risk that consumers will experience premature spoilage due to spore-forming bacteria.
MATERIALS AND METHODS
On-Farm Intervention
Details of the intervention strategy and spore count testing methods can be found in the previous study conducted by
. Briefly, an on-farm spore intervention was applied on 5 New York State conventional dairy farms and included: (1) adding chlorine bleach to the wash cycle for the laundered towels used during udder preparation before milking unit attachment, (2) drying the laundered towels completely following the wash cycle, and (3) training milking parlor employees on the importance of thorough teat end cleaning during milking. All parts of the intervention were applied concurrently on each farm, and farms were sampled 3 times over 15 mo. Farms were sampled on a rotating basis so that 1 farm was sampled each month and each farm had 4 mo between samplings. For each implementation, samples were collected every other day for 1 wk before the intervention was applied and every other day for a week after the intervention was applied (
, was used in this study to predict how the shelf life of fluid milk was affected by varying levels of initial raw milk contamination with spores of psychrotolerant spore-forming bacteria. Details of this baseline model development and validation are outlined in
), or version 1 of the model, was modified in 2 ways: (i) for simulated half-gallon samples with an initial count <1 cfu/half-gallon (i.e., a fraction of a bacterial cell, which is not realistic), version 1 allowed bacterial growth, whereas the current version, version 2, did not allow bacterial growth on any day in shelf life when the initial contamination level was below 1 cfu/half-gallon; (ii) in version 1, all simulated half-gallon samples were held at the same constant refrigeration temperature during shelf life (i.e., constant of 6°C), whereas for version 2, the temperature for shelf life storage assigned to simulated half-gallon samples was instead drawn for each iteration (n = 10,000; 10 simulated half-gallons per iteration all had the same storage temperature) from a Laplace (mean = 4.06°C, SD = 2.31°C) distribution as modeled in a 2005 consumer study on refrigeration temperatures and food storage habits (
). The temperature distribution used in version 2 of the model more closely estimated real consumer refrigerated storage conditions, and unlike version 1 of the model, it did not assume that every consumer holds their milk at a static temperature (6°C) throughout its shelf life.
Our baseline, pre-intervention model used the parameters (e.g., initial spore levels, spore-forming bacteria subtypes) as reported in
), the initial log most probable number spore counts were sampled from a normal distribution with a mean of −0.723 log cfu/mL for version 2. To evaluate the effect of implementing the on-farm intervention, a 0.22 log reduction was applied to this distribution. This reduction value was based on the reduction of mesophilic spores observed in
, with an assumption that the reduction of psychrotolerant spores would be similar to the reductions observed for mesophilic and thermophilic spores. The model included an assumption that there were upper and lower limits for refrigeration temperatures (minimum = −1°C, maximum = 15°C). Monte Carlo simulations comprised 10,000 iterations. All model programming and statistical analyses were performed in R (version 3.6.3; R project) and can be found at https://github.com/FSL-MQIP/FISE.
Cost of Intervention by Farm Size
The cost analysis assumed farms will apply a combination of all the interventions tested in the previous study conducted by
(e.g., washing towels with bleach, drying towels, and giving a onetime training to milking parlor employees on the importance of teat end cleanliness) at the same time. Assumptions were made when calculating the cost of the intervention, including: (1) 1 dryer can dry 15 1-h loads of 100 towels per day, (2) the cost of a dryer is $1,000, (3) the average lifespan of a dryer is 3 yr, (4) the cost of electricity is $0.12/kWh, (5) the average dryer uses 3.3 kW of energy per hour, (6) a 3-pack of concentrated bleach (Clorox) costs $14.15, (7) milking employee training costs $125 per year, (8) cows are milked 3 times per day, and (9) farms average milk yields of 12,700 kg/cow per year (28,000 lbs/cow per year). Yearly costs were calculated based on herd size. Since 4 of the 5 farms in the study conducted in
dried their towels before the start of the study, an additional analysis was done to calculate the cost of the intervention if the farm did not need to purchase a dryer or pay for additional electricity to apply the intervention. This analysis estimated the costs of an intervention that modifies milking towel management that already includes drying to add a chlorine treatment during washing. Additional potential costs not considered here, but that may affect some producers, include increases in labor costs for producers who do not dry towels (e.g., to account for the amount of additional time it would take to move the towels from the washer to the dryer, and so on), although these costs are likely to be minimal.
The training costs were based on training prices set by Quality Milk Production Services at Cornell University for 1 h of milking parlor employee training for an out-of-state farm. These costs do not include employee pay for the duration of the training (i.e., 1 h). As typical employee wage rates vary greatly based on parlor style, farm size, farm location, and other farm management practices, these costs would need to be considered by individual producers implementing the intervention.
Consumer Survey Design and Distribution
A survey to evaluate consumer home use of fluid milk was designed on Qualtrics (Qualtrics) and was distributed electronically through the Cornell University Food Science Sensory Evaluation Center as well as through social media. The survey was developed to determine consumer fluid milk consumption patterns, specifically, the date printed on the container (i.e., code date) and consisted of 25 questions with 495 respondents who completed the survey (Supplemental Table S1, https://github.com/FSL-MQIP/FISE). Briefly, survey questions fell into the following categories: (1) basic demographic information; (2) type of milk consumption consumed by each household member; and (3) milk consumption habits.
RESULTS AND DISCUSSION
Significant Reductions in Bulk Tank Raw Milk Spore Levels Can Be Achieved with a Low-Cost, On-Farm Intervention
A farm-level, spore intervention strategy that included washing towels with bleach, drying towels, and providing a onetime training to milking parlor employees, was found in a previous study to significantly reduce bulk tank spore levels (
). Our analysis determined that the cost to implement the intervention ranged between $9.49 and $13.35 per cow per year based on herd size (Table 1), or between $0.03 and $0.05 per hundred-weight (CWT) of raw milk depending on herd size, assuming a herd average of 12,700 kg (28,000 lbs) produced per cow per year (Table 1). According to the 2019 New York State Dairy Business Summary (NYSDBS), which evaluated 153 dairy farms across the state, the cost of applying this intervention is low when compared with the overall costs associated with maintaining a dairy cow each year (e.g., feed, bedding, veterinary care, and so on) (
). Specifically, the average of all expenses associated with maintaining a milking cow (i.e., not calves or heifers) in New York State in 2019 for the 147 farms that participated in the NYSDBS in both 2018 and 2019, including the cost of feed, bedding, milking supplies, machinery, utilities, veterinary care, and hired labor was $5,040 per cow per year and the average cost of maintaining a milking cow per CWT of milk produced was $19.16 (
). Furthermore, 4 of the 5 farms who participated in the previous study using this intervention already used dry towels before they were used during milking preparation (
), meaning that they were already paying for the cost of purchasing a dryer and the associated electricity. For farms that already dry towels, the estimated intervention cost ranged from an additional $4.45 to $5.64 per cow per year depending on the size of the farm or will cost about $0.02 per CWT, assuming a herd average of 12,700 kg (28,000 lbs) produced per cow per year (Table 1).
Table 1Cost of on-farm spore intervention per year by cow and by hundred-weight (CWT) of milk based on herd size
Item
Factors affecting cost of spore interventions by herd size
Importantly, our analysis indicates that the cost per cow of applying the intervention strategy drops as the herd size increases. Others have described how economies of scale, or the cost advantages attained as a business grows, can save dairy farmers money. For instance,
). Mosheim and colleagues suggest that this is because the opportunity costs of hired labor is high for small dairies and because small dairies are less likely to adopt new technologies that will make them more efficient due to the high cost per cow (
In addition to the outcome of significantly reducing the bacterial spore level in bulk tank raw milk, we hypothesized that this on-farm intervention strategy would also reduce the risk of mastitis and high somatic cell counts, as the intervention focuses on cleaner teat ends and good milking parlor practices, which would provide additional cost savings to producers. Several studies have found that cleaner teat ends are associated with fewer cases of mastitis (
). For example, de Pinho Mazi and colleagues conducted a study to evaluate the relationship between teat end condition and mastitis and found that cows with poor teat end condition scores were 30% more likely to develop mastitis than cows with good teat end condition scores, and cows with dirtier udders were 47% more likely to develop mastitis than cows with clean udders (
). Understanding that mastitis can be reduced by having cleaner udders and teats is important because mastitis is one of the most expensive costs to a dairy farm. Producers not only have to pay to treat the cow with mastitis and withhold the milk for either high somatic cell count or for antibiotics if she is treated, but the cow's lifetime milk yield potential, or the amount of milk a cow could produce in her lifetime, is reduced (
Holland, J., J. Hadrich, C. Wolf, and J. Lombard. 2015. Economics of measuring costs due to mastitis-related milk loss. Selected paper prepared for presentation at the 2015 AAEA & WAEA Joint Annual Meeting, San Francisco, CA.
involves milking parlor employee training on the importance of teat end cleanliness during milking, we hypothesized that the intervention may also reduce the incidence of mastitis on a farm, although field studies will need to be performed to test this hypothesis and quantify the impact of this intervention on mastitis frequency. Even small reductions in mastitis frequency, however, would be expected to make this intervention cost neutral, as the costs of mastitis to dairy farmers are substantial and have been estimated to average $179 for a single clinical mastitis case, though it varies greatly based on cow traits such as age at first clinical mastitis case and stage of lactation as well as the current milk price (
, will reduce spoilage in HTST pasteurized fluid milk. Fluid milk was used as a model system to evaluate the influence of reduced spore raw milk on pasteurized product shelf life, as it has been reported that spore-forming bacteria are the predominant organism present in milk that exceeds the PMO limit of 20,000 cfu/mL for approximately 50% of fluid milk in New York (
Pseudomonas fluorescens group bacterial strains are responsible for repeat and sporadic postpasteurization contamination and reduced fluid milk shelf life.
on the shelf life of HTST fluid milk. The model indicated that 17 d after pasteurization, representing the approximate average number of shelf-life days for HTST fluid milk, 18.24% of simulated half-gallon samples had bacterial counts above 20,000 cfu/mL with an overall mean of 1.73 log cfu/mL without an intervention applied, while with the intervention applied, 16.99% of simulated half-gallon samples had bacterial counts above 20,000 cfu/mL with an overall mean of 1.53 log cfu/mL (Table 2). This represents a 6.85% reduction in half-gallon containers of milk with bacterial counts above 20,000 at 17 d postpasteurization (Figure 1). Our model predicted that by 21 d after pasteurization, 27.90% of half-gallon samples had bacterial counts above 20,000 cfu/mL without the intervention applied, while 26.30% of half-gallon samples exceeded that level when raw milk from farms implementing the intervention was simulated (Table 2).
Table 2Monte Carlo simulation model predicted mean spore counts and percent of fluid milk samples with spore counts >20,000 cfu/mL with and without interventions applied
Whether or not the on-farm intervention was applied. The intervention included washing towels with bleach, drying towels, and giving a one-time training to milking parlor employees on the importance of teat end cleanliness.
Mean spore count (log cfu/mL)
Samples >20,000 cfu/mL (%)
10
No
0.36
4.56
10
Yes
0.15
4.15
14
No
1.15
11.76
14
Yes
0.94
10.95
17
No
1.73
18.24
17
Yes
1.53
16.99
21
No
2.44
27.90
21
Yes
2.25
26.30
24
No
2.92
35.64
24
Yes
2.74
33.65
28
No
3.48
45.77
28
Yes
3.32
43.67
31
No
3.84
52.52
31
Yes
3.69
50.40
35
No
4.24
60.04
35
Yes
4.12
58.40
1 Number of days after HTST pasteurization.
2 Whether or not the on-farm intervention was applied. The intervention included washing towels with bleach, drying towels, and giving a one-time training to milking parlor employees on the importance of teat end cleanliness.
Figure 1Histograms representing the Monte Carlo simulated distribution of bacterial spore counts in fluid milk (a) 17 d and (b) 21 d after pasteurization for original milk samples without the intervention applied (red bars) and for the postintervention samples with the intervention applied (blue bars). The intervention included washing towels with bleach, drying towels, and giving a once-a-year training to milking parlor employees on the importance of teat end cleanliness.
While these reductions in simulated fluid milk reaching the PMO bacterial limit by the end of typical product shelf life are small, they represent a meaningful contribution to the overall goal of reducing consumer exposure to product with reduced quality. Importantly, incremental reduction in microbial spoilage of fluid milk reduces the risk that consumers will be exposed to product with reduced quality. However, we acknowledge that our model is based off the PMO bacterial limit of 20,000 cfu/mL while spoilage that can be sensorially perceived typically only occurs in milk with >1 million cfu/mL (
on higher value dairy products such as dairy powders, which are often subjected to strict spore specifications.
A survey of 495 consumers was conducted to illuminate the fluid milk consumption patterns of HTST fluid milk consumed in their homes. Overall, 92% (456/495) of respondents indicated that they or someone they live with consumes milk, and from that subset, 95% (435/456) consume milk themselves. Of the respondents who indicated that they consume milk, 55% (249/456) reported that they continue to drink milk past the date printed on the container (often referred to as code date or best-by date). The majority, or 54% (111/207), of those who do not drink milk past the date printed on the container responded that this is because they always finish their milk before the date, while 42% (87/207) indicated that they do not drink milk past the code date and will discard their milk when it reaches the date printed on the container. A typical fluid milk code date is approximately 17 d after pasteurization, a point where our model showed that there is a small but significant reduction in the number of simulated fluid milk in half-gallon containers exceeding the PMO bacterial limit when the low-cost, on-farm spore intervention is applied. This reduction would reduce the number of instances of premature spoilage or reduced quality fluid milk that over half of the milk-drinking population is exposed to and will have the added benefit of reducing consumer food waste, which is a major issue in the United States.
The implications of consumer dissatisfaction with food product quality is highlighted by a study by Quelch and Ash, in which consumers were surveyed regarding their satisfaction with food products (
). The authors found that the top 2 reasons for consumer dissatisfaction with fresh foods, including fluid milk, were “The product was spoiled, defective, or damaged” and “The quality was poorer than expected.” Further, while only 2% of consumers indicated that they complain to the manufacturer in response to unsatisfactory purchase experiences with food products, 14.7% report complaining to the store, which could influence whether or not a store continues to carry that product (
). Importantly, 19.6% of consumers decided to not buy that brand again and 11.6% of consumers stated that they warn their friends and family about the product (
may represent an underestimation of the impact of consumer dissatisfaction with food products as the study was conducted over 4 decades ago, before the advent of social media, which has been shown to amplify consumer complaints (
). Due to the far-reaching impacts of consumer dissatisfaction with food products, it is critical for dairy industry stakeholders to reduce the risk of consumer exposure to reduced quality fluid milk products using strategies such as those outlined here, including implementing a low-cost, on-farm spore intervention.
CONCLUSIONS
Applying a low-cost, easy-to-implement intervention on dairy farms will not only reduce psychrotolerant spores in the bulk tank raw milk but may also extend fluid milk shelf life, reduce exposure of consumers to reduced quality product, and reduce the environmental impact of dairy by reducing food waste. Our study provides dairy producers with key information to assess the economic benefits of spore reductions and will provide data for processors to set premiums for low-spore milk. Ultimately, dairy industry stakeholders across the fluid milk continuum are negatively impacted by consumer dissatisfaction with milk products, including spoilage and reduced quality resulting from the growth of spore-forming bacteria originating from raw milk, highlighting the need for dairy processors to incentivize the implementation of this intervention through premiums to producers. Future studies should focus on the effectiveness of this intervention and its impact on the incidence of mastitis, in addition to developing models that can predict the impact this intervention would have on cheeses and dairy powders.
ACKNOWLEDGMENTS
This project was funded by the New York Farm Viability Institute (OSP#78590; Syracuse, NY) and the consumer survey was funded by the Foundation for Food and Agriculture Research (FFAR, Washington, DC, award no. CA18-SS-0000000206). We also extend our greatest appreciation to the New York State dairy producers and their employees who participated in this study. We recognize the staff and students of the Milk Quality Improvement Program at Cornell University (Ithaca, NY) who dedicated their time and energy to this project. The authors have not stated any conflicts of interest.
REFERENCES
Alles A.A.
Wiedmann M.
Martin N.H.
Rapid detection and characterization of postpasteurization contaminants in pasteurized fluid milk.
Holland, J., J. Hadrich, C. Wolf, and J. Lombard. 2015. Economics of measuring costs due to mastitis-related milk loss. Selected paper prepared for presentation at the 2015 AAEA & WAEA Joint Annual Meeting, San Francisco, CA.
in: De Vos P. Garrity G.M. Jones D. Krieg N. Ludwig W. Rainey F.A. Schleifer K. Whitman W.B. Bergey’s Manual of Systematic Bacteriology. 3rd ed. Springer US,
2009: 21-128
Pseudomonas fluorescens group bacterial strains are responsible for repeat and sporadic postpasteurization contamination and reduced fluid milk shelf life.
00031-0&id=gr1.jpg){kind=link}