Ranking disease control strategies with stochastic outcomes

This paper explains how the methodologies of first and second order stochastic dominance, and expected utility using specific risk preferences, can be applied to epidemiology when choosing among control strategies that have stochastic outcomes. We provide a step-by-step guide on how epidemiologists can rank a number of control strategies based on their distribution of estimated benefits. We also explain how the expected utility model and decision maker's risk preferences can be used to select between outcomes when none stochastically dominates. To illustrate these techniques, we show the ranking of various control strategies for a dairy herd endemically infected with Mycobacterium avium subs. paratuberculosis (MAP) and mastitis, and explain how decision maker's risk preferences affect the ranking.

A decision-support tool for investment analysis of automated oestrus detection technologies in a seasonal dairy production system

Context Advances in automated oestrus detection have made this an attractive technology to help reduce manual oestrus detection labour on dairy farms. Aims A decision-support tool was created to help farmers estimate the investment outcome of adopting automated oestrus detection technologies in a seasonal dairy production system. Methods A decision-support tool was created using Excel 2011 (Microsoft Inc., Redmond, WA, USA). The tool allows farmers to input both current herd reproductive management costs and performance and automated oestrus detection technology system costs and performance to receive herd-specific estimates of investment benefit. The investment analysis outputs include the net present value (NPV), internal rate of return (IRR), and payback period associated with automated oestrus detection adoption. Two different automated oestrus detection technologies were compared with visual oestrus detection aided by tail paint with a 72.0% oestrus detection rate (sensitivity) to demonstrate the value of the investment analysis tool. The alternative scenarios, technology one and technology two, were compared over an eight-year investment period. Key results Technology one, with a 62.4% oestrus detection rate, resulted in a negative NPV and IRR (–NZ$182567 and –100% respectively), indicating a poor investment. Technology two, with an oestrus detection rate of 91.0%, provided a positive NPV and IRR (NZ$177890 and 38.7% respectively), indicating a beneficial investment. The payback period for technology one was estimated as >10 years, whereas technology two’s payback period was <1 year. Conclusions The investment tool results are dependent on farm-specific and automated oestrus detection inputs. Implications Farmers can use farm-specific inputs in the tool to aid them when considering adoption of new automated oestrus detection technologies.

Economic considerations of breeding for polled dairy cows versus dehorning in the United States

This paper examines the costs and benefits of selecting for polled dairy heifers versus traditional dehorning practices. Stochastic budgets were developed to analyze the expected costs (EC) associated with polled dairy genetics. The economic assessment was expanded beyond on-farm cash costs by incorporating cost and benefit estimates to generate industry-wide discussion, and preliminary economic evaluations, surrounding the public acceptance and attitude toward polled genetics versus dehorning calves. Triangular distributions, commonly used to represent distributions with limited data, were used to represent labor costs for dehorning, the likelihood of treatment of calf, and the cost of veterinary treatment. In total, 10,000 iterations were run using @Risk v 6.0 (Palisade Corp., Newfield, NY). The EC of the 4 traditional dehorning methods evaluated in this study ranged from $6 to $25/head, with a mean EC around $12 to $13/head. The EC of incorporating polled genetics into a breeding program ranged from $0 to $26/head depending on the additional cost, or premium, associated with polled relative to horned genetics. Estimated breakeven premiums associated with polled genetics indicate that, on average, producers could spend up to $5.95/head and $11.90/head more for heterozygous and homozygous polled genetics, respectively, compared with conventional horned genetics (or $2.08 and $4.17/straw of semen at an assumed average conception rate of 35%). Given the parameters outlined, sensitivity to individual farm semen and dehorning costs are likely to swamp these differences. Beyond on-farm costs, industry-wide discussion may be warranted surrounding the public's acceptance and attitude toward polled genetics versus dehorning or disbudding of calves. The value of avoiding dehorning may be larger for the industry, and perhaps some individual farms, than initially suggested if additional value is put on calf comfort and possible worker aversion to dehorning. If public perception of dehorning influences market access, the EC of dehorning may be large but that cost is unknown at present.

Stochastic economic evaluation of dairy farm reproductive performance

Kalantari, A. S. and Cabrera, V. E. 2015. Stochastic economic evaluation of dairy farm reproductive performance. Can. J. Anim. Sci. 95: 59–70. The objective of this study was to assess the economic value of reproductive performance in dairy farms under uncertain and variable conditions. Consequently, the study developed methods to introduce stochasticity into transition probabilities of a Markov chain model. A robust Markov chain model with 21-d stage length and three state variables, parity, days in milk, and days in pregnancy, was developed. Uncertainty was added to all transition probabilities, milk production level, and reproductive costs. The model was run for 10 000 replications after introducing each random variable. The expected net return (US$ cow−1 yr−1±standard deviation) was $3192±75.0 for the baseline scenario that had 15% 21-d pregnancy rate (21-d PR). After verifying the model's behavior, it was run for 2000 replications to study the effect of changing 21-d PR from 10 to 30% with one-unit-percentage interval. The economic gain of changing 21-d PR from 10 to 30% resulted in a US$75 cow−1 yr−1, and this overall increase in the net return was observed mainly due to the lower reproductive and culling cost and higher calf value. The gain was even greater when milk price and milk cut-off threshold decreased.

Economic impacts of adoption and fundraising strategies in animal shelters

The adoption strategies used in animal shelters can have a large impact on the total number of adoptions and donations that take place. Reducing adoption fees during peak kitten or puppy season is one way to reduce inventories and increase the number of open spaces to save more lives, but does not necessarily increase the financial well-being of the shelter if the per-animal costs exceed the revenues generated. We developed a stochastic model to simulate the expected costs, revenues, and net income of a hypothetical animal shelter for various alternative management strategies, based on US conditions. A total of 8 scenarios were developed and compared to the base-case scenario (BC). In the model, scenarios which decreased or waived adoption fees caused total costs to increase due to the escalating costs associated with increasing the total number and density of animals housed. This effect was especially pronounced when adoptions were free. When the return on money invested in additional fundraising was predetermined to be 'good' (rather than 'fair' or 'poor'), net shelter income did exceed costs - but even 'fair' return increased net shelter income compared to the BC. Of the eight scenarios compared to BC, the mean monthly net income was significantly different from that in the BC in all eight scenarios (p<0.01). In contrast, variances were different (p<0.01) in five of the eight scenarios (and the uncertainty that comes with high variance would make planning difficult for shelter managers); however, the variance in net income did not differ from the BC for any of the scenarios investigating returns to additional spending on promotion and fundraising. In these scenarios, because the extra cost involved is relatively low compared to the other scenarios, the potential risk of a reduction in net shelter revenue is reduced. When shelters are aware of the positive and negative impacts of various adoption strategies on mean net income and variation in net income, shelter managers can better strategize saving animal lives and meeting shelter goals, while maintaining the financial health and functionality of the operation.

Invited review: sensors to support health management on dairy farms

Since the 1980s, efforts have been made to develop sensors that measure a parameter from an individual cow. The development started with individual cow recognition and was followed by sensors that measure the electrical conductivity of milk and pedometers that measure activity. The aim of this review is to provide a structured overview of the published sensor systems for dairy health management. The development of sensor systems can be described by the following 4 levels: (I) techniques that measure something about the cow (e.g., activity); (II) interpretations that summarize changes in the sensor data (e.g., increase in activity) to produce information about the cow's status (e.g., estrus); (III) integration of information where sensor information is supplemented with other information (e.g., economic information) to produce advice (e.g., whether to inseminate a cow or not); and (IV) the farmer makes a decision or the sensor system makes the decision autonomously (e.g., the inseminator is called). This review has structured a total of 126 publications describing 139 sensor systems and compared them based on the 4 levels. The publications were published in the Thomson Reuters (formerly ISI) Web of Science database from January 2002 until June 2012 or in the proceedings of 3 conferences on precision (dairy) farming in 2009, 2010, and 2011. Most studies concerned the detection of mastitis (25%), fertility (33%), and locomotion problems (30%), with fewer studies (16%) related to the detection of metabolic problems. Many studies presented sensor systems at levels I and II, but none did so at levels III and IV. Most of the work for mastitis (92%) and fertility (75%) is done at level II. For locomotion (53%) and metabolism (69%), more than half of the work is done at level I. The performance of sensor systems varies based on the choice of gold standards, algorithms, and test sizes (number of farms and cows). Studies on sensor systems for mastitis and estrus have shown that sensor systems are brought to a higher level; however, the need to improve detection performance still exists. Studies on sensor systems for locomotion problems have shown that the search continues for the most appropriate indicators, sensor techniques, and gold standards. Studies on metabolic problems show that it is still unclear which indicator reflects best the metabolic problems that should be detected. No systems with integrated decision support models have been found.