Livestock and poultry play a vital role in the upliftment of the rural poor farmers by improving their income and employment. Expanding urbanization and increasing per capita incomes, especially in countries like India, are continuously pushing up the demand for protein rich livestock and poultry products, which in turn resulted in increasing intensification and commercialization of these subsectors to augment output. Together with the paradigm shift in the farming system, the genetic makeup of the animals and birds has undergone a laudable shift, thus improving the production and productivity of livestock industry. Despite the tremendous increase in livestock and poultry output in the last six decades, this rosy picture is not without any issues. When the country started importing improved germplasms, the exotic pure and crossbreds could not withstand the hardships and contingencies peculiar to Indian agroclimate and started suffering from many new diseases / disorders, which started affecting the profitability of the farm business. Further, a disease in livestock not only brings down the attainable revenue within the farm, but also has serious negative effects on the related industries like feed and products processing businesses. Most importantly, the presence of contagious diseases like Foot and Mouth Disease has adverse havoc on the export markets. Unfortunately, there are no adequate reading materials on the economics of animal diseases and their control. Although developed countries had started working on estimating animal disease costs and disease control and prevention costs in the early 1970s, developing countries like India initiated research works in this field only in 1990s, that too sporadically. This book has been written with the basic purpose of documenting the principles and methods of animal health economics, relatively a newer discipline for the country, to propagate the knowledge of the concepts and procedures in measuring animal disease frequencies, disease loss estimation methods, and economics of disease control programmes to help the practicing veterinarians, academicians, research workers, planners and administrators of livestock and poultry sectors.

Human life on the earth has been intertwined with the fabric of crops and livestock for its sustenance, since time immemorial, for food and protection against the vagaries of nature. When agricultural and livestock production systems began to change in the 1950s, from traditional extensive systems to semi-intensive and intensive production systems, diseases of rural farm enterprises became challenging. Especially, the emerging diseases in livestock continued to demand more and more investments for ensuring food safety and security. Despite significant progress made in the recent past in reducing the prevalence of animal diseases, there is still an increasing concern over the losses associated with diseases of economic importance that result in production inefficiency. The scientific foundation for the discipline of animal health economics was laid by Morris (1969) and further steps were taken by Ellis (1972) to economically evaluate the Swine Fever eradication programme in Great Britain using cost-benefit analysis, Martin Hugh-Jones (1975) to specialize in GIS to model diseases and assessed the economic impact on production by infectious diseases, James and Ellis (1978) to use the same tool for FMD control programme, Renkema and Stelwagen (1979) to evaluate the replacement rates in dairy herds, Renkema (1980) to analyze the economic aspects of diseases in animals, Ellis and Putt (1981) to suggest the epidemiological and economic implications of FMD vaccination programme in Kenya, and McInerney (1983) to analyze the economics of animal diseases in dairy cattle, the multidisciplinary field of animal health economics was not amply exploited especially in developing countries to assess the extent of livestock disease losses (Ngategize and Kaneene, 1985).

Statistics is defined as a discipline that includes procedures and techniques used to collect, process, and analyze numerical data to make inferences about the data. The word statistic is used as a singular, denoting the numerical data. Statistics are the numerical quantities calculated from sample observations. Here, the word statistics hence is plural. Basically, statistics is defined as the statistical data (in the plural sense) and also as statistical methods (in the singular sense). A clearer and more precise definition is that ‘statistics is the study of methods and procedures for collecting, classifying, presenting, analyzing, and interpreting data to make scientific inferences from it’. Bio statistics is the science dealing with the application of statistical methods to the problems of biological sciences. 2.1 FUNCTIONS OF STATISTICS a) To present the facts in a definite form: If the data or results are given in numbers, rather than in qualitative statements, they will be more useful. The statements like ‘there is a considerable amount of unemployment in a country or the human population in a state is increasing at a faster rate’ are not in definite forms. The statements should be in definite form like there is a 15 percent unemployment in 2023 or the population is increasing by 5 percent every decade.

3.1 MEASURES OF CENTRAL TENDENCY (AVERAGES) When data are presented in raw form or in the form of frequency distribution, they fail to provide a general idea of the data under study. The general idea of statistical methods is to condense the raw data into a form that can convey some meaningful characteristic(s) of the data. Though classification simplifies the data and improves the understanding of the main characteristics in the data, it is still difficult to remember. In any distribution of data, values of a variable may tend to cluster (or concentrate) around a central value. The data must be condensed into a single value. Such presentation of data is called a Central Tendency and the measures devised to consider this tendency are known as Measures of Central Tendency. An average condenses a frequency distribution or raw data and presents it in a single representative number. It is that single value which is considered as the most representative or typical value for a set of values. Such a value usually lies somewhere near the centre of the group. That is why, an average is usually referred to as a measure of central tendency. It is located at a point around which most of the other values tend to cluster and therefore it is also termed as a measure of central location.

For any statistical analysis, the required primary data are usually collected by the census or sampling methods. In the census method, the data are collected from all the individuals. E.g., If we want to study the weight of 100 cows, we have to make use of all 100 cows for data collection. That is, we collect data from the entire population or universe. However, in the sampling method, the data are collected from a small group of the population which is termed as the sample. A sample is a portion of the population selected to represent the population. E.g., we collect data only from a sample of cows, say 20 cows drawn from the population.

Before trying to understand Hypothesis Testing and Experimental Designs, we should be conversant with the concept of Probability Distribution. 5.1 PROBABILITY DISTRIBUTIONS Let us look at an example to grasp the essence of a probability distribution. Let us rotate a pair of dice and observe the sum by totalling the numbers that turn upon in both the dice. Let us call this sum by a random variable X. When the first dice shows up number 1, the second dice could show 1, 2, 3, 4, 5, or 6. Likewise, when the first dice shows up number 2, the second dice could show 1, 2, 3, 4, 5, or 6. If we continue the experiment this way until all the possibilities are exhausted, the sample space will contain 36 elements. The distribution that can be observed will contain a sum = 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12. The starting sum will be = 2, because when both the dice show the number 1, the sum 1 + 1 = 2. Likewise, the maximum sum in the distribution will be 12, when both the dice show up a number 6. If we write the values of the random variable (sum) with associated probabilities, it is called a probability distribution.

A scientific experiment is a planned research to obtain new facts or to confirm or refute the results of previous experiments. Research is usually conducted in a controlled environment to study the effect of one or more variables on observations. The objective of research may be to answer a question, verify a hypothesis, or estimate the effect of an independent variable on a dependent variable. After stating the research problem, we need to outline the methodology (or materials and methods) of the research, which includes developing an experimental design. Scientific experiments are conducted to study the causal relationships between independent variable(s) and dependent variable(s). The researcher manipulates one or more independent variables and measures their effect on one or more dependent variables. Designing an experiment is planning an experiment. It is a complete sequence of steps taken before experimenting to ensure that appropriate data are obtained to perform a data analysis to get valid inferences about the problem. It must be seen that the data collected are relevant to the problem investigated. E.g., studying the effect of breed, age, feed, etc. on milk output. Here, the objects like breed, age, feed, etc. are called treatments. That is, any specific experimental conditions (like the addition of feed additives, new feed formula, etc.) applied to the experimental units are called treatments. The basic objects on which the experiment is done are called experimental units (E.g., a bird, an insect, an animal). However, the responses in all the experimental units receiving the same treatments may not be similar, due to the inherent differences in the experimental units, called experimental error, caused by extraneous factors.

Veterinary epidemiology is the study of disease within a livestock population, dealing with the investigation (including identification, quantification, and examination), of the direct and indirect determinants of disease distribution in animal populations. Any epidemiological investigation starts with whether there is an animal health problem in a population, which requires data collection on animal, management, economic, and environmental factors. These data are then analyzed descriptively to assess the frequency of disease occurrence. The results are then used to decide whether investigations of causal relationships between disease occurrence and potential risk factors are necessary for the development of control or prevention policies. Once the relationships are understood, necessary interventions for the disease are then planned and their impacts assessed. Hence, the goal of epidemiological investigations is the control and prevention of animal disease problems by integrating the available pieces of evidence into the decision-making process for disease control policy development.

8.1 NEED FOR ANIMAL DISEASE FREQUENCY MEASURES The very essence of veterinary epidemiology is to measure animal disease occurrence and to make comparisons between the animal population groups. The animal disease frequency measures help us to understand the distribution of diseases in animal populations. Hence, the principal role of veterinary epidemiology is to describe and explain differences in the distribution of animal diseases between the populations. Measures of the disease frequencies are used to describe how common an illness is concerning the size of the population (the population at risk) at a particular point in time. The major reason for determining the morbidity is to assess the extent of a disease problem in a population. This assessment is possible only when the number of diseased animals is calculated and compared with the total number of animals in the population. A report of 10 cases of mastitis in a dairy herd does not indicate the true extent of the disease problem, unless the report also indicates the total number of dairy animals in the herd or population. Thus, it is important to relate the number of cases of a disease to the population at risk of developing that disease.

9.1 ECONOMIC LOSSES DUE TO ANIMAL DISEASES Economic considerations are increasingly forcing their way forward in the decision-making processes of animal health management. With the continuing intensification of livestock production systems, the economic implications of livestock diseases have become increasingly important at both farm and national levels, as they represent avoidable waste of scarce resources (McInerney, 1988). From an economic point of view, an animal disease either destroys the basic resources, reduces the attainable output, dwindles the efficiency with which the resources are converted into products, or decreases the economic value of products (McInerney, 1995). The effects of a disease present in an animal or herd may be manifested by stunted growth, lowered production, delayed maturity, delayed conception, mortality, reduced resource (especially feed) use efficiency, etc. These negative impacts of a disease demand more resource(s) than usual, resulting in higher expenditures for the same output and reduced farm income. Apart from these negative effects at the farm or herd level, animal diseases lead to economic losses at the regional, national, and global levels too by increasing animal mortality, reducing productivity, increasing disease control costs, swelling losses in animal trade, decreasing the market values of livestock and their products, and ultimately ending at food insecurity. These economic impacts of livestock diseases, besides their social impacts, have now been recognized globally, in both developed and developing countries, necessitating quantification of the economic impact of animal diseases as an urgent and important step to support the prevention and control decisions for improved animal health to ensure better rural and global economy.

After assessing the prevalence or incidence of the disease concerned, the next step is to quantify the economic losses due to that disease by classifying and presenting all the information available on the losses, expressing these in monetary values, and identifying and quantifying the indirect losses attributable to a disease, to the extent possible. Usually, most of the negative effects of diseases can be conveniently estimated in terms of reduced output, mortality, reduced weight, delayed maturity, and treatment cost, directly. However, quantifying the indirect effects of the disease will be complex, but it is not impossible. As we saw earlier, the economic impacts of animal diseases can be measured using statistical / epidemiological methods and economic methods. Economic methods of disease loss estimation include the Gross Margin Analysis, Enterprise Budgeting, Equi-Marginal Principle, Partial Budgeting, Cost Benefit Analysis, and Decision Analysis.

An animal disease is an economic process in the farming system, as it consumes a considerable amount of scarce economic resources that would have been otherwise productively utilized by the animals and results in negative benefits to society. Hence, while measuring the losses attributable to animal diseases, the manifestation of such losses and the factors influencing such losses may also be identified, which require modelling the ways by which the negative effects of the diseases are exhibited. A description of such models and their procedures is essential to understand the comprehensive effects of diseases. We will list and describe such models and tools in this chapter. 11.1 CORRELATION ANALYSIS While simple descriptive statistical tools are adequate to estimate the economic effects of animal diseases, these tools cannot measure the nature and strength of the relationships between two or more variables. However, the statistical knowledge can effectively be used to understand Various statistical measures studied so far, like measures of central tendency and dispersion and skewness, relate to one variable only. However, there are many situations where there are several problems involving the use of two or more variables. Although correlation analysis is only a bivariate technique like t-test, ANOVA, and Chi square test, we will briefly see about correlation in this chapter, as it is also a part of multivariate analysis. The first step is to determine which bivariate analytical tool we need to adopt to test the study hypothesis.

Programming techniques are used the mathematical tools by the scientists, economists, and planners to specify the optimum combination of resources and enterprises in farms and firms, to identify the profit maximizing mixes of commodities to be produced, to specify the cost minimizing methods of production and transportation of goods, to identify the least cost locations from which raw materials can be purchased, to determine the locations of production at which a particular commodity should be produced and the consumption centres to which it should be shipped, and to specify the least cost ingredients of a feed or ration which must meet the specified nutritional requirements. Programming techniques can be either linear or non-linear. Although several non-linear programming techniques (like quadratic, integer, separable, dynamic programming, etc., which are very complex to work with) are available, we will largely deal with linear programming technique and to a certain extent will brief the theory of dynamic programming.

There are three types of disease occurrence namely, endemic, epidemic and outbreaks. Martin et al. (1993) described the term endemic as the disease occurrence in a predictable pattern. There can be low, moderate or high levels of endemicity. Although many times the terms epidemic and outbreak are used synonymously, the term epidemic is used when there is an unexpected incidence of a disease. Thrusfield (1995) defined an outbreak as the occurrence of a disease involving one or more animals. Animal disease prevention and control activities offer valuable benefits for food production, food security and safety, animal and human welfare, and alleviation of poverty. The effectiveness of disease prevention and control policies depends on the good governance and quality of the veterinary services, in compliance with OIE standards and guidelines on animal disease control. The apex body in each country is the core system that takes care of the prevention and control of animal diseases. This body is responsible for early detection of diseases and rapid response to outbreaks. The veterinary services in developing countries require adequate human and financial resources and capacity building to carry out the tasks of disease prevention and control and to protect animal health and thus public health, including food security and food safety.

14.1 NEED FOR ANIMAL DISEASE INFORMATION SYSTEMS Once the animal disease data are gathered, a few things need to be done with the data so collected. First, they must be managed and quality checked; Secondly, they must be analyzed to be understandable; and thirdly, they must be acted upon. For data to be analyzable and useful for decision-making, they need cautious management and quality control. An information system is the collection of data, people, procedures, hardware, software, files, and data required to accomplish an organized set of functions. The constituents of an information system are: • People – those who gather data, those responsible for data input and those responsible for data analysis • A storage / retrieval / analysis system – which consists of computer hardware and appropriate software • A feedback delivery system – mechanisms to ensure that processed data are fed back to data gatherers

Abelson, P. (2008), Establishing a monetary value for lives saved: issues and controversies, Working papers in CBA 2008-02. Paper prepared for the conference on Delivering better-quality proposals through better CBA, Office of Best Practice and Regulation, Department of Finance and Deregulation, Government of Australia. Ahuja, V., R. Rajabova, D. Ward and A. McLeod (2009), Willingness to pay for disease prevention: Case of brucellosis control in Khatlon Oblast of Tajikistan, Report of the European Commission Project, ‘Enhancing individual incomes and improving living standards in Khatlon, Tajikistan’, Reference: EuropAid/125-743/L/ACT/TJ. Ahuja. V. and A. Sen (2006), Willingness to pay for veterinary services: Evidence from poor areas in rural India. PPLPI Research Report RR Nr. 06-03. Rome, FAO. Atkinson, A.C. (1990), A note on the generalized information criterion for choice of a model, Biometrika, 67: 413 – 418. Awa, D.N., A. Njoya and A.C. Ngo Tama (2000), Economics of Prophylaxis against PPR and gastrointestinal helminthiasis in small ruminants in North Cameroon, Tropical Animal Health and Production, 32: 391 – 403.