Preventive Veterinary Medicine 113 (2014) 1–12
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Economic losses due to cystic echinococcosis in India: Need for urgent action to control the disease Balbir B. Singh a,∗ , Navneet K. Dhand b , Sandeep Ghatak c , Jatinder P.S. Gill a a
School of Public Health & Zoonoses, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana, Punjab 141004, India Faculty of Veterinary Science, The University of Sydney, 425 Werombi Road, Camden, 2570 NSW, Australia c Division of Animal Health, ICAR Research Complex for the North Eastern Hill Region, Umroi Road, Umiam, Barapani, Meghalaya 793103, India b
a r t i c l e
i n f o
Article history: Received 31 March 2013 Received in revised form 22 July 2013 Accepted 14 September 2013 Keywords: Hydatidosis Echinococcosis Economic loss India Developing countries Parasitic zoonosis
a b s t r a c t Cystic ehinococcosis (CE) caused by Echinococcus granulosus remains a neglected zoonotic disease despite its considerable human and animal health concerns. This is the first systematic analysis of the livestock and human related economic losses due to cystic echinococcosis in India. Data about human cases were obtained from a tertiary hospital. Human hydatidosis cases with and without surgical interventions were extrapolated to be 5647 and 17 075 per year assuming a total human population of 1 210 193 422 in India. Data about prevalence of hydatid cysts in important food producing animals were obtained from previously published abattoir based epidemiological surveys that reported a prevalence of 5.39% in cattle, 4.36% in buffaloes, 3.09% in pigs, 2.23% in sheep and 0.41% in goats. Animal population data were sourced from the latest census conducted by the Department of Animal Husbandry, Dairying and Fisheries, India. Other input parameters were obtained from published scientific literature. Probability distributions were included for many input values to account for variability and uncertainty. Sensitivity analyses were conducted to evaluate the effect of important parameters on the estimated economic losses. The analysis revealed a total annual median loss of Rs. 11.47 billion (approx. US $ 212.35 million). Cattle and buffalo industry accounted for most of the losses: 93.05% and 88.88% of the animal and total losses, respectively. Human hydatidosis related losses were estimated to be Rs. 472.72 million (approx. US $ 8.75 million) but are likely to be an under-estimate due to under-reporting of the disease in the country. The human losses more than quadrupled to Rs. 1953 million i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased to 0.2% in the sensitivity analyses. The social loss and psychological distress were not taken into account for calculating human loss. The results highlight an urgent need for a science based policy to control and manage the disease in the country. © 2013 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding author. Tel.: +91 161 2414009; fax: +91 161 2400822. E-mail address: [email protected] (B.B. Singh). 0167-5877/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.prevetmed.2013.09.007
Cystic echinococcosis (CE) is an important zoonosis caused by Echinococcus granulosus affecting both the livestock and human populations, particularly in developing countries such as India (Singh et al., 2010). Dogs act as a definitive host of E. granulosus; humans and food
2
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
producing animals acquire the infection by ingestion of eggs shed by the definitive host (Thompson, 1995). Cysts of E. granulosus have been found in a variety of intermediate and aberrant hosts such as cattle, buffalo, sheep, goat, pigs, horses, camel and man (Rausch, 1986, 1995; Thompson and Allsopp, 1988; WHO, 1984). In livestock, CE may lead to reduced yield and quality of meat, milk and wool; reduced birth rate; delayed performance and growth; and condemnation of organs, especially of liver and lung (Benner et al., 2010; Budke et al., 2005; Torgerson et al., 2000; Polydorou, 1981). Economic losses also occur due to destruction of infected viscera and dead animals or due to ban on export of animals and their products if these are required to be free of CE. Human echinococcosis continues to be a significant public health problem in several regions of the world. Human losses primarily occur due to the need for surgical interventions, hospital care and productivity losses in asymptomatic cases. A recent Spanish study reported high economic losses at D 148 964 534 with human disease constituting most (89.1%) of the total losses (Benner et al., 2010). In Chile, human hydatidosis was associated with the average mortality rate of 0.2 deaths per 100 000 inhabitants per annum and a loss of 3349 years of life due to the premature death of 235 people (Martinez, 2011). The rural population of India is dependent on livestock for their livelihood (Singh et al., 2011) where cattle and buffalo production systems constitute the backbone of the livestock industry. Hydatid cysts have been found in most of the food producing animals in India such as cattle, buffalo, sheep, goat and pigs. Presence of stray animals such as dogs and cattle, unorganised slaughtering, free access of dogs to slaughter waste, decline in vulture population, sewage irrigation, and improper garbage disposal methods increase the chances of disease transmission to humans (Singh et al., 2002). Despite being a significant zoonotic problem, not many studies have been conducted in India to understand disease epidemiology or to estimate economic losses due to the disease. As a result, virtually no programs have been developed or implemented at the national level to control or eradicate the disease. This is the first systematic analysis of the economic losses due to CE in India. Such studies are required to help develop disease prevention and control strategies for India and other developing countries. 2. Methods 2.1. Livestock related losses The important food producing animals that make a substantial contribution to the meat and dairy industry in the country—sheep, goat, cattle, buffalo and pigs—were included in the analysis. The prevalence of CE in livestock was based on a recent abattoir based epidemiological survey conducted in India (Singh, 2011). The official data for animal populations (Table 1) were obtained from the latest census carried out by the Department of Animal Husbandry, Dairying and Fisheries, India (DAHD & F, 2010). The annual wool output per sheep was taken from published scientific literature (Banerjee, 1991) while the average milk
Table 1 Cost parameters used to estimate the economic losses associated with cystic echinococcosis in livestock, India. Average cost, in Rs. per kg
Reference
Sheep Sheep carcass
131.26
Sheep liver
131.26
Ranjan and Rawat (2011) Ranjan and Rawat (2011) Market value NABARD (2010)
Parameter
Sheep lung Sheep wool
40 38
Goats Goat carcass
131.26
Goat liver
131.26
Goat lung Goat’s milk at farm gate Cattle Beef carcass Cow liver Cow lung Cow’s milk Buffalo Buffalo carcass
40 20 112.18 60 35 26
Ranjan and Rawat (2011) Market value Market value Market value
Buffalo liver Buffalo lung Buffalo’s milk
60 35 30
Ranjan and Rawat (2011) Market value Market value Market value
Pigs Pig carcass Pig liver Pig lungs
95 50 30
Wright et al. (2010) Market value Market value
Energy (kW h)
112.18
Ranjan and Rawat (2011) Ranjan and Rawat (2011) Market value Market value
2.82
Punjab State Electricity Regulatory Commission (2010) Punjab (India)
yield and carcass weight were sourced from the official data (DAHD & F, 2010). The drought power production equivalence was taken from previous studies (Banerjee, 1991). The price of energy (kW h) charged by Punjab State Electricity Regulatory Commission (2010) Punjab (India) was used to calculate losses in draught power. The other related input parameters such as organ weight, animal produce, productivity losses (decrease in fecundity, carcass weight, milk production, draught power and wool output), life expectancy and reproductive rates were obtained from the published scientific literature (Table 2). The prices of animal carcasses, milk, liver and lungs were sourced through market surveillance or from published scientific literature (Table 1). Both direct and indirect economic losses for all the species were calculated separately. While direct losses included those occurring due to offal condemnation of infected animals, the indirect losses included those occurring due to reduction in growth and production and decrease in fecundity. All the analyses were conducted using R-statistical program (R-statistical package version 2.12.0, R Development Core Team, http://www.r-project.org).
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
3
Table 2 Epidemiological parameters used to estimate the economic losses associated with cystic echinococcosis in livestock, India, 2007. Parameter Sheep Population Total population Total breedable female population No. of sheep slaughtered per year Prevalence of infection At inspection In liver In lungs Production Sheep carcass Sheep liver Sheep lung Wool per sheep Mean lambing per year per ewe No. unborn lambs Lamb mortality rate Goats Population Total population Total breedable female In milk No. of goats slaughtered per year Prevalence of infection At inspection In livers In lungs Production Average carcass weight Goat liver Goat lung Mean kidding per year per doe Average milk yield of goat Lactation length No. unborn kids Kid mortality Cattle Population Total population Total breedable female population In milk No. of slaughtered cattle per year Prevalence of infection At inspection In livers In lungs Production Average carcass weight Cattle liver Cattle lung Mean calving per year per cow Average milk yield of cattle Drought power production equivalence Calf mortality No. of unborn calves Average working hrs per year per drought cattle Buffalo Population Total population In milk Breedable female population No. of slaughtered Buffalo per year Prevalence of infection At inspection In livers In lungs
Value
Range
Unit
Distribution
Reference
71 558 000 53 684 000 23 269 000
NA NA NA
Individual Individual Individual
Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
2.23 1.44 1.32
NA NA NA
% % %
Beta (18, 742) Beta (12, 748) Beta (11, 749)
Singh (2011) Singh (2011) Singh (2011)
13 566.5 623.5 1.0 1.25 82 304 10
453–680 340–907 NA 0.9–1.6 NA NA
kg g g kg/year Individual Individual %
Fixed Uniform Uniform Fixed Uniform Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) Banerjee (1991) Calculation NABARD (2010)
140 537 000 62 489 000 28 868 000 50 707 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
0.41 0.28 0.24
NA NA NA
% % %
Beta (11, 2428) Beta (8, 2431) Beta (7, 2432)
Singh (2011) Singh (2011) Singh (2011)
10 566.5 623.5 2.67 0.37 160 37 624 15
NA 453–680 340–907 2.01–3.33 NA NA NA NA
kg g g Individual kg/day day Individual %
Fixed Uniform Uniform Uniform Fixed Fixed Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Banerjee (1991) Calculation NABARD (2010)
199 075 000 72 915 000 38 928 000 2 476 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
5.39 4.68 4.32
NA NA NA
% % %
Beta (16, 262) Beta (14, 264) Beta (13, 265)
Singh (2011) Singh (2011) Singh (2011)
90 5.4 2.6 1 4.505 0.75
NA NA 2.2–3.0 NA 2.14–6.87 NA
kg kg kg Individual kg/day HP
Fixed Fixed Uniform Fixed Uniform Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Banerjee (1991)
5 216 157 360
NA NA NA
% Individual h
Fixed Fixed Fixed
NABARD (2010) Calculation Calculation
105 343 000 35 479 000 54 475 000 5 884 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
4.36 4.36 4.36
NA NA NA
% % %
Beta (14, 284) Beta (14, 284) Beta (14, 284)
Singh (2011) Singh (2011) Singh (2011)
4
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Table 2 (Continued) Parameter Production Average carcass weight Buffalo liver Buffalo lung Mean calving per year Average milk yield of dairy buffalo No. of unborn calves Calf mortality Pigs Population Total population No. of pigs slaughtered Total no. of breedable female Prevalence of infection At inspection In livers In lungs Production Average weight of pig carcass Pig lung Pig liver Mean no. of piglets per year No. of unborn pigs Piglet mortality Productivity losses—all livestock Decrease in fecundity Decrease in carcass weight Decrease in milk production Decrease in drought power output Decrease in wool output Productive lifespan Farmer’s benefit from unborn animal produce
Value
Range
Unit
Distribution
Reference
106 5.4 2.6 1 4.57 130 631 5
NA NA 2.2–3.0 NA NA NA NA
kg kg kg Individual kg/day Individual %
Fixed Fixed Uniform Fixed Fixed Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Calculation NABARD (2010)
11 134 000 6 746 000 3 055 000
NA NA NA
Individual Individual Individual
Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
3.09 2.82 1.97
NA NA NA
% % %
Beta Beta Beta
Singh (2011) Singh (2011) Singh (2011)
35 396.5 1.45 16 83 072 20
NA 340–453 0.9–2 NA NA NA
kg g kg Individual Individual %
Fixed Uniform Uniform Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) Banerjee (1991) Calculation NABARD (2010)
5.5 6.25 2.5 2 2.5 35 15
0.0–11.0 2.5–10.0 0.0–5.0 NA NA NA 10–20
% decrease per year % decrease per year % decrease per year % decrease per year % decrease per year % of average lifespan %
Triangular Triangular Triangular Triangular Triangular Fixed Uniform
Benner et al. (2010) Benner et al. (2010) Benner et al. (2010) Benner et al. (2010) See text See text See text
2.1.1. Direct losses The direct losses occurring due to offal condemnation were calculated based on average weight and market price for liver and lung for all the five species taking into account the infected slaughtered animal population in India. For example, the direct losses due to condemned lungs for cattle were calculated as follows: Number of infected condemned lungs = Number of cattle slaughtered per year × Prevalence of infection in lungs
2.1.2.1. Losses due to reduction in milk production. To estimate losses due to reduction in milk yield in the cattle, buffalo and goat industry, we first calculated the number of infected ‘in milk’ animals during the year (Eq. (3)) and their lactation yield (Eq. (4)) from the official data provided by Department of Animal Husbandry, Dairying and Fisheries, India (DAHD & F, 2010). Annual milk losses were then calculated by taking into account the reduction in milk yield due to CE (Eqs. (5 and 6)) and the price of milk (Eq. (7)). The losses due to decrease in milk yield for the other two dairy species were calculated in a similar manner.
(1) Infected ‘in milk’ animals = Total number of cows in milk × Prevalence of infection
(3)
Losses due to condemned infected lungs = Number of lungs condemned × Average weight of lung × Price of lung
(2)
Milk produced per animal per lactation = Average milk production per animal per day
Similarly, the losses due to condemnation of liver and lungs were estimated for all the species. The total direct losses were calculated by combining losses from all species. 2.1.2. Indirect losses Decrease in milk production, growth and fecundity of infected animals were the important indirect losses considered in this study.
× Lactation length (days)
(4)
Milk loss per animal per annum due to CE = Milk produced per animal per lactation × Proportional reduction in milk production due to CE per year
(5)
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Total milk loss due to CE = Milk loss per animal due to CE × Infected in milk animals
(6)
Cost of lost milk = Total milk loss due to CE × Price of milk (7)
2.1.2.2. Losses due to reduction in carcass weight. To calculate losses due to reduced carcass weight at the time of slaughter, first the number of infected animals slaughtered during the year were calculated (Eq. (8)), followed by reduction in carcass weight per animal (Eqs. (9 and 10)). Annual losses due to reduced carcass weight were then estimated by including the price of meat (Eq. (11)). The Eqs. (8–11) show calculation of indirect losses due to reduced carcass weight for one species; losses for other species were also calculated in a similar manner. No. of infected animal slaughtered per year
5
To estimate economic losses, we estimated the numbers of males and females in these infected and uninfected groups but equations are only shown for the healthy offspring (Eq. (17)). Of the unborn male offspring that would have been slaughtered or used for draught purposes were calculated as shown in Eqs. (18–19). The losses from unborn male offspring without CE were then estimated for meat (Eqs. (20–21)) and for drought power (Eqs. (23–25)). Similarly, losses from female offspring were calculated as shown in Eqs. (26–28). No. of infected breedable females = Total breedable female population × Prevalence of infection
(12)
No. of unborn offspring = No. of infected breedable females × Mean numbers of offspring per year per animal × Decrease in fecundity
(13)
= No. of animal slaughtered per year × Prevalence of infection at inspection
(8)
No. of unborn offspring that would have survived = Unborn offspring
Loss of carcass weight per animal due to CE
× (1 − Newborn offspring mortality rate)
(14)
= Average weight of carcass × Percentage reduction in carcass weight
(9)
= Unborn offspring that would have survived
Total loss of carcass weight
× Prevalence of infection
= No. of infected animal slaughtered per year × Loss of carcass weight per animal due to CE
(10)
(15)
No. of healthy unborn offspring without CE = Unborn offspring that would have survived
Cost of lost carcass weight = Total loss of carcass weight × Price of meat
No. of unborn offspring infected with CE
(11)
2.1.2.3. Losses due to reduction in fecundity. Reduction in fecundity was calculated from infected breedable population (Eq. (12)) and the annual number of offspring (DAHD & F, 2010) as described by Benner et al. (2010). These calculations were based on the assumptions that prevalence of infection was equal in breeding and non-breeding populations. First we calculated the number of infected breedable females (Eq. (12)) and then the number of unborn offspring from them due to reduction in fecundity (Eq. (13)). However, not all of these unborn offspring would have survived because some of them would have died due to other reasons. This was accounted for in Eq. (14) and then the number of infected and uninfected unborn offspring that would have survived were estimated (Eqs. (15–16)) after assuming an equivalent prevalence of infection in unborn animals.
− Unborn offspring infected with CE
(16)
No. of unborn (male/female) offspring without CE = Healthy unborn offspring without CE × Percentage of (male/female) offspring born
(17)
No. of unborn male offspring without CE that would have been slaughtered = Unborn male offspring without CE × Proportion of male animals that undergo slaughter
(18)
Unborn male offspring without CE that would have been used for draught = Unborn male offspring without CE × Proportion of male animals used for draught purpose
(19)
6
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Losses due to decrease in fecundity
Productive lifespan = Average lifespan × Percentage of lifespan in production
(20)
= Losses from healthy unborn offspring without CE + Losses from healthy unborn offspring with CE
(30)
Meat loss due to CE = Healthy unborn male offspring without CE that would have been slaughtered × Average weight of carcass
Actual loss in farmers’ profit due to decrease in fecundity (21)
× Farmer’s profit percentage
Cost of lost meat = Meat loss due to CE × Price of meat (22) Draught energy produced per animal over lifetime = Av draught power equivalence per animal
(31)
The indirect losses due to decrease in fecundity for other species were calculated in similar fashion with slight modifications. For example, sheep wool losses were calculated as follows: Wool produced per animal over lifetime
× Working hours per year per animal × Productive lifespan
= Total loss due to decrease in fecundity
(23)
= Productive lifespan × Average wool produced per animal per year
(32)
Total loss of draught power Wool loss due to CE = Healthy unborn offspring without CE
= Draught energy produced per animal over lifetime × Healthy unborn male offspring without CE that would have been used for draught purposes
× Wool produced per animal over lifetime
(33)
(24) Cost of lost wool = Wool loss due to CE × Price of wool
Cost of lost draught power =
(34)
Total loss of draught power 1000
× Price of draught energy/kW h
(25)
Milk produced per animal over lifetime = Milk produced per animal per lactation × Productive lifespan
(26)
Total milk loss from unborn healthy female offspring = Milk produced per animal over lifetime × Healthy unborn female offspring without CE
(27)
Cost of lost milk = Total milk loss from unborn healthy female offspring × Price of milk
(28)
Total losses were calculated by combining losses from meat, drought power and milk (Eq. (29)). Losses from unborn offspring infected with CE were also calculated in the similar manner. The total losses due to decrease in fecundity were calculated as shown in Eqs. (30 and 31). Losses from healthy unborn offspring without CE = Cost of lost meat + Cost of lost drought power + Cost of lost milk
(29)
Finally, overall losses were calculated from decreased fecundity, meat and wool losses in sheep; meat and milk losses in goat; meat, milk and draught power losses in cattle; milk and meat losses in buffalo and meat losses in pigs. The buffalo industry in India does not have any distinct meat and dairy breeds. In this industry, females are reared for milk and males for meat purposes. Therefore, the calculations were done assuming equal populations of unborn male and female animals. The scenario of the cattle industry in the country is different due to the ban on cow slaughtering in most parts of the country. The female animals are reared for milk but the fate of male animals is not fully known. Therefore, to calculate annual losses due to decrease fecundity in cattle industry, we assumed that all of the female population was reared for milk, 5% of the male population was slaughtered and 30% of the male population was used for draught power due to partial mechanization of Indian agriculture. As per the official estimates, slaughtered cattle population were approximately 3.5% of the total male populations. However, to adjust for unaccounted or illegal slaughtering practices, we assumed that 5% of the male population was slaughtered. Similarly, the male population used for other purposes except for breeding was 60% of the total male population. We assumed that at least half of that population (30%) is used for drought purposes. As cattle bulls are used in agricultural operations in India, the decrease in draught power for draught cattle was also estimated. After calculating lifetime costs of animal produce and draught power from unborn animals, we assumed that
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
7
Table 3 Epidemiological and economic parameters used to estimate the human economic losses associated with cystic echinococcosis in India. Parameter
Value
Distribution
Reference
Total number of diagnosed cases without surgery Total number of diagnosed cases with surgical/interventional procedure Undiagnosed or asymptomatic cases Productivity loss as % per year Cost of diagnosed cases without surgery Cost of diagnosed cases involving surgical/interventional procedures Per capita income (per annum)*
17 075 5646
Fixed Fixed
Khurana et al. (2007) Khurana et al. (2007)
0.00–0.021–0.04 0–4 Rs. 2000 Rs. 50 000
Triangular Uniform Fixed Fixed
See text Benner et al. (2010) Personnel communication Personnel communication
Rs 33 283
Fixed
1 210 193 422
Fixed
Ministry of Statistics and Programme Implementation, Govt. of India Census of India, 2011
Total population *
For the year 2007–2008.
farmer would receive 15% (range 10–20%) benefits from rearing of these animals after subtracting all the other investment costs. This assumption was used to calculate actual loss in farmer profits due to decrease in fecundity. 2.2. Human related losses 2.2.1. Sources of data The human epidemiological parameters used in the analysis included the prevalence of diagnosed cases with surgical intervention, diagnosed cases without surgical intervention and undiagnosed or asymptomatic cases. To estimate the productivity losses in humans, the number of reported CE cases was obtained from analysis of the retrospective data of a tertiary hospital (Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India) (Khurana et al., 2007). On the average, 1 318 271 patients visited the hospital per year during 2007–2009 (42nd annual report, PGIMER). During the past 20 years 495 cases were diagnosed with CE of which 123 required surgical intervention and 372 were treated without surgical intervention. The total numbers of diagnosed cases in the country were extrapolated from the PGIMER data (Table 3). 2.2.2. Estimation of costs for treating cases The fixed costs were calculated for two types of cases: treatment of diagnosed cases who had gone for surgery to remove cysts (Eq. (35)) and treatment of patients who were diagnosed but had not gone for removal of cysts (Eq. (36)).
2.2.3. Estimation of costs due to loss of productivity Losses due to productivity were estimated for three types of cases: (a) diagnosed cases requiring surgery; (b) diagnosed cases without surgery; and (c) undiagnosed cases. To calculate productivity losses, official per capita income per annum was obtained and productivity losses due to CE were taken from published scientific literature (Eq. (37)). The productivity losses associated with diagnosed cases with and without surgical intervention were estimated as shown in Eqs. (38–39). Loss in annual per capita income due to CE = Annual per capita income × Annual productivity percentage loss
(37)
Loss in annual per capita income due to CE from diagnosed cases without surgery = Loss in annual per capita income due to CE × Total number of diagnosed cases without surgery
(38)
Loss in annual per capita income due to CE from diagnosed cases with surgical or interventional procedure = Loss in annual per capita income due to CE × Total number of diagnosed cases with surgical or
Losses due to treating CE diagnosed cases with surgery
interventional procedure
(39)
= Total number of diagnosed cases with surgical or interventional procedure × Cost of treating diagnosed cases involving surgical or interventional procedures
(35)
Losses due to treating CE diagnosed cases without surgery = Total number of diagnosed cases without surgery × Cost of treating diagnosed cases without surgery
(36)
The productivity losses associated with undiagnosed cases were not possible to be directly estimated because undiagnosed or asymptomatic cases have never been identified in India using ultrasound. Therefore, the prevalence of these cases was calculated based on a similar study carried out in Florida (Uruguay), where prevalence of undiagnosed or asymptomatic cases was found to be 1.64% by ultrasound, with annual surgical incidence of 36.1 cases per 105 person-years, and a ratio of 45.4 (Benner et al., 2010; Carmona et al., 1998). First, incidence rate of surgical cases in India was calculated to be 0.466 cases per
8
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
105 person-years based on the total cases requiring surgical intervention (5646). From this estimate, the mean prevalence of undiagnosed or asymptomatic CE cases was calculated as 0.021%. These calculations were based on comparison between annual surgical incidence in India with prevalence of undiagnosed or asymptomatic cases by ultrasound and annual surgical incidence in Florida (Uruguay).
Mean prevalence of undiagnosed or asymptomatic CE cases in India
=
Prevalence of undiagnosed or asymptomatic cases in Florida
Annual surgical incidence per 105 person years in Florida
× Annual surgical incidence per 105 person years in India
(40)
2.4. Sensitivity analyses Sensitivity analyses were conducted to evaluate the effect of changing input values of prevalence of undiagnosed asymptomatic cases and per capita income (per annum) for which objective information was not available. Budke et al. (2006), assumed that ≈10% of annual cases are not officially diagnosed due to socioeconomic status of the patients or the subclinical nature of the illness. Therefore sensitivity analysis was conducted to make adjustments for under-reporting. We assumed a very low proportion of asymptomatic cases; therefore, it was increased substantially in the sensitivity analysis. Similarly, we changed per capita income to evaluate the effect of income on losses due to the disease. 3. Results
Total annual loss in per capita income due to CE from undiagnosed cases = Loss in per capita income per year due to CE × Total number of undiagnosed cases
(41)
2.3. Accounting for variability and uncertainty
In total, CE was found to cause median economic losses to the amount of Rs. 11.47 billions (approx. US $ 212.35 million) (95% CI 6.63–19.48 billion) per annum (Table 4). Median livestock and human related losses amounted to Rs. 10.95 billion (95% CI 6.11–18.74 billion) and 0.472 (95% CI 0.330–0.737 billion), respectively, comprising 95.52% and 4.12% of the total losses (Fig. 1). 3.1. Livestock related losses
Beta probability distributions were used to account for uncertainties in the prevalence of CE. Alpha (˛) and beta (ˇ) parameters of beta probability distributions were based on the recent abattoir based epidemiological survey conducted in India (Singh, 2011).
The economic losses related to cattle, buffalo, sheep, goat and pigs are provided in Table 4. Occurrence of CE in cattle and buffalo accounted for 93.05% of the total livestock and 88.88% of the total (livestock and human) losses.
˛ = number of positive CE cases + 1.
3.2. Human related losses
ˇ = total number of cases examined − number of positive CE cases + 1. Uniform distributions were used for weight of the condemned offal. The minimum and maximum values for weight of the condemned liver and lungs were taken from the published scientific literature (Gracey and Collins, 1992). Triangular distributions were applied for decrease in fecundity, carcass weight, milk production, draught power, wool output. For calculating actual farmer profits due to decrease in fecundity, we applied uniform distribution in the range of 10–20% because benefits are likely to vary under different rearing conditions. We used a triangular distribution with a mode of 0.021% and a range of 0.00–0.04 for the proportion of undiagnosed or asymptomatic human cases for analysis of economic losses. Uniform distributions were applied to calculate productivity loss (0.0–4.0%) in diagnosed and undiagnosed cases (Torgerson et al., 2000). The similar distributions were also applied in a Spanish study (Benner et al., 2010). The confidence limits for the estimates were calculated and running Monte Carlo simulations for 10 000 iterations.
CE was found to cause median human losses amounting Rs. 472.7 million (approx. US $ 8.75 million). 3.3. Sensitivity analysis The sensitivity analysis revealed that annual median losses were Rs. 496.9 million (95% CI 334–652 million) (approx. US $ 9.20 million) at the human undiagnosed prevalence of 0.02 (Fig. 2). However, the annual median losses increased to Rs. 1953 million (95% CI 488–3413 million), i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased 10 times to 0.2%. The sensitivity analysis further revealed that human losses were Rs. 362.9 million (95% CI 320.4–442 million), i.e. approx. US $ 6.72 million, assuming the annual per capita income of Rs. 10 000 (Fig. 2). However the losses rose to Rs. 781.6 million (95% CI 358.2–1578 million), i.e. approx. US $ 14.47 million when the annual per capita income of Rs. 100 000 was assumed. 4. Discussion As far as we are aware, this is the first systematic analysis of economic losses occurring due to CE in India. The results indicate that CE causes huge economic losses in India at the estimated prevalence rates of the
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
9
Table 4 Total losses associated with cystic echinococcosis in humans and livestock in India estimated in the study. Total losses
Indian rupees Median
Sheep Goat Cattle Buffalo Pig Human Total animal loss Grand total loss
113 787 069 53 274 837 5 599 822 658 4 592 145 524 96 141 417 472 715 931 10 952 980 207 11 467 000 000
US dollars 95% confidence intervals 75 875 331–167 202 569 32 850 991–84 511 258 2 133 969 312–12 588 894 415 2 071 978 664–9 275 390 197 55 561 238–1 59 243 769 330 337 589–737 059 556 6 105 879 161–18 736 319 358 6 629 764 108–19 478 312 127
Median 2 107 168 986 571 103 700 420 85 039 732 1 780 397 8 753 999 202 832 967 212 351 852
95% confidence intervals 1 405 099–3 096 344 608 352–1 565 023 39 517 950–233 127 674 38 369 975–171 766 485 1 028 912–2 948 959 6 117 363–13 649 251 113 071 836–346 968 877 122 773 409–360 709 484
Fig. 1. Total, livestock and human associated losses estimated in the study due to cystic echinococcosis in India.
10
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Human loss (Million rupees)
a) Human undiagnosed cases 4000 3500 3000 2500 2000 1500 1000 500 0
0.02
0.05
0.1
0.15
0.2
Human loss (Million rupees)
Proporon of undiagnosed cases b) Per capita income
1800 1600 1400 1200 1000 800 600 400 200 0
10
20
30
40
50
60
70
80
90
100
Per capita income (thousand rupees) Fig. 2. Sensitivity analysis of a) asymptomatic human undiagnosed cases and b) annual per capita income to account for losses associated with human cystic echinococcosis. The tick marks indicate the median losses and the bars extend to the 95% lower and upper confidence intervals.
infection. Globally, Budke et al. (2006) estimated human annual loss of US $ 194 million (US $ 171–217 million). When they accounted under-reporting, the losses increased up to US $ 763 million (US $ 676–857 million). An annual livestock production loss of at least US $ 141 million (US $ 101–183 million) and possibly up to US $ 2190 million (US $ 1572–2951 million) was also estimated across the globe (Budke et al., 2006). High costs were estimated in several other countries also (Torgerson et al., 2001; Moro et al., 2011; Benner et al., 2010; Fasihi et al., 2012). All these results indicate that CE causes substantial economic losses across the globe and India is no exception. Estimation of economic losses of a zoonosis gives best indication of the combined human and livestock sector losses (Carabin et al., 2005). From the livestock sector, cattle and buffalo industries contributed most of the losses due to CE. Decrease in fecundity in cattle and buffalo represented the biggest indirect loss associated with CE. These losses could have serious consequences as dairy industry is the largest and most important livestock industry in India. However, the losses due to slaughter of infected cattle were fairly low. This disparity is due to religious and cultural beliefs in India. Cows are considered to be holy animals in India, particularly among Hindus, which constitute about 80% of the Indian population. Therefore, cow slaughter is banned in most of the Indian states (Singh et al., 2013). The cattle are mainly reared for milk production and upon death, the carcasses are usually left for scavenging. The figure of 2.47 million for the number of cattle slaughtered per year reflects only the cattle that are slaughtered each year. Occurrence of major losses in the livestock industry indicates that a greater
emphasis should be placed on controlling CE in animals particularly in cattle and buffalo. While indirect losses were a major component of the livestock related losses, the direct losses were significant for human CE associated economic losses. The human losses were influenced by prevalence of undiagnosed cases and more than quadrupled to Rs. 1953 million i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased to 0.2% in the sensitivity analyses. Similarly, per capita income was also found to have substantial impact on human related economic losses. Additionally, CE causes huge psychological and socioeconomic losses which could not be accounted in the present study. Such losses could arise due to permanently decreased quality of life and a loss of job (Torgerson, 2003). We acknowledge that the human related prevalence rates of infection are likely to be an under-estimate because not everyone has access to advanced medical care facilities in the country, leading to under-reporting of the human diseases. Secondly, the asymptomatic or undiagnosed cases are likely to be an under-estimate as ultrasonography based surveys have not been conducted in the country and the prevalence of asymptomatic or undiagnosed cases was calculated from the studies carried out in other countries. Therefore, human related cost calculations are lower compared to economic losses due to human CE as reported across the globe (Budke et al., 2006) and in many other countries such as Spain (Benner et al., 2010), Peru (Moro et al., 2011), Jordon (Torgerson et al., 2001) and Iran (Fasihi et al., 2012). The disease control programmes usually rely on public education, upgrading of meat inspection facilities, dog population control and active detection and elimination of E. granulosus in infected dogs. However, presence of many factors such as huge stray dog population, unorganised slaughtering, free access of dogs to slaughter waste (Singh et al., 2013), decline in vulture population (Markandya et al., 2008) and routine contamination of pastures with dog faeces make it really difficult to control the disease in India. Holistic implementation of animal birth control programme at the country level to control the ever increasing stray dog populations is a must for India and many other developing countries so as to break the transmission cycle of the disease. Many of the costs assumed in our analysis for medical care were consensual fixed costs but are in fact likely to vary in different parts of the country. For example, we used estimates of human prevalence from a hospital in northern India and extrapolated it to the whole population of India. These input parameters could be biased but the detailed epidemiological studies on human hydatidosis have not been conducted in southern India. Many case reports of human hydatidosis from central and southern India (Rao et al., 2012; Akther et al., 2011; Tiwary and Tiwary, 1988; Mehta et al., 1982; Parija et al., 1983; Vamsy et al., 1991) indicate that the disease is endemic in these parts of the country as in northern India. Due to similar socioeconomic, hygienic and animal husbandry management practices followed throughout the country, we expect a similar prevalence in these parts of the country.
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Similarly, many of the costs assumed for animal produce were also fixed. Although, we used different prices of meat to calculate losses due to infected lungs and liver but the same prices were used to calculate losses due to rest of the carcasses for all the animals. This was realistic because most of the slaughterhouses in the country sell carcasses at gross prices. We believe that these limitations will not cause much difference to the total calculated losses but a small over- or under-estimate could not be ruled out. The study had a number of strengths. To obtain valid estimates of economic losses due to animals and humans, we used official figures for most of the parameters. For the input parameters where official data were unavailable, data from previously published research were used. For many input parameters, we used probability distributions already used in the previous studies to account of variability and uncertainty (Benner et al., 2010; Budke et al., 2006; Torgerson et al., 2000, 2001; Heath et al., 2003; Baitursinov et al., 2004). However, we could not consider many potentially influential factors due to unavailability of data. For example, we used overall prevalence of infection for calculating production losses. The size and number of the cysts were not taken into account for calculating such losses because this information was not available. Similarly, the human losses were estimated based on national annual per capita income. It has certain disadvantages as it does not take into account informal labour such as family care giving. Moreover, the prevalence of infection could vary in people belonging to different economic classes and could lead to an over or under-estimate of economic losses. However, we applied triangular distributions to account for uncertainty in losses arising due to decrease in fecundity, carcass weight, milk yield, wool output and drought power. Moreover, we conducted sensitivity analyses to evaluate the influence of input parameters such as prevalence of undiagnosed asymptomatic cases and per capita income (per annum) for which objective information was not available. Similarly, other factors which could influence indirect livestock losses such as animal breed, diet and management conditions could not be taken into consideration. We also did not estimate decrease in hide value due to animals infected with CE. We also did not include partial compensation through replacement for losses due to unborn animals, as most of the farmers keep only a few animals and do not have enough resources to replace for their unborn animals. Within the constraints of the study, we believe that livestock and human CE losses calculated in the present study represent the human and livestock related losses occurring due to CE in India. However, further studies should be conducted to refine the models and obtain precise information about input parameters from various parts of the country. 5. Conclusion CE is an important zoonosis affecting most of the food producing animals and man in India. The economic analysis revealed heavy losses associated with CE in the dairy industry. High prevalence of the disease in the cattle and buffalo were found to be important factors for huge economic losses. Human losses are likely to be an under-estimate due
11
to under-reporting of the disease in the country. There is urgent need for action based on implementation of intervention policies such as control of stray animal populations, prophylactic anthelmintic treatment of dogs, organised slaughtering and use of proper garbage disposal systems. Acknowledgements This study was financially supported by the School of Public Health & Zoonoses, Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana, Punjab, India. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j. prevetmed.2013.09.007. References Akther, M.J., Khanam, N., Rao, S., 2011. Clinico epidemiological profile of hydatid diseases in Central India, a retrospective and prospective study. Int. J. Biol. Med. Res. 2 (3), 603–606. Banerjee, G.C. (Ed.), 1991. A Text Book of Animal Husbandry. , Seventh ed. Oxford & IBH Publishing Pvt. Ltd, Kolkata. Baitursinov, K.K., Shaikenov, B., Abdybekova, A.M., 2004. Distribution of Echinococcus amongst agricultural animals in Kazakhstan. In: Torgeson, P.R., Shaikenov, B. (Eds.), Echinococcosis in Central Asia: Problems and Solutions. Dauir Publishing House, Almaty, Kazakhstan, pp. 101–118. Benner, C., Carabin, H., Serrano, L.P.S., Budke, C.M., Carmena, D., 2010. Analysis of the economic impact of cystic echinococcosis in Spain. Bull. World Health Organ. 88, 49–57. Budke, C.M., Deplazes, P., Torgerson, P.R., 2006. Global socioeconomic impact of cystic echinococcosis. Emerg. Infect Dis. 12 (2), 296–303. Budke, C.M., Jiamin, Q., Qian, W., Torgerson, P.R., 2005. Economic effects of echinococcosis in a disease endemic region of the Tibetan Plateau. Am. J. Trop. Med. Hyg. 73 (1), 2–10. Carabin, H., Budke, C.M., Cowan, L.D., Willingham, A.L., Torgerson, P.R., 2005. Methods for assessing the burden of parasitic zoonoses: echinococcosis and cysticercosis. Trends Parasitol. 21, 327–333. Carmona, C., Perdomo, R., Carbo, A., Alvarez, C., Monti, J., Grauert, R., 1998. Risk factors associated with human cystic echinococcosis in Florida Uruguay: results of a mass screening study using ultrasound and serology. Am. J. Trop. Med. Hyg. 58, 599–605. DAHD & F, 2010. Basic Animal Husbandry Statistics. Department of Animal Husbandry, Dairying & Fisheries, Ministry of Agriculture, Government of India. Fasihi, H.M., Budke, C.M., Rostami, S., 2012. The monetary burden of cystic echinococcosis in Iran. PLoS Negl. Trop. Dis. 6 (11), e1915. Gracey, J.F., Collins, D.S., 1992. Meat Hygiene, Ninth ed. Bailliere Tindall, England. Heath, D.D., Jensen, O., Lightowlers, M.W., 2003. Progress in control of hydatidosis using vaccination—a review of formulation and delivery of the vaccine and recommendations for practical use in control programmes. Acta Trop. 85, 133–143. Khurana, S., Das, A., Malla, N., 2007. Increasing trends in seroprevalence of human hydatidosis in North India: a hospital based study. Trop. Doct. 37, 100–102. Markandya, A., Taylor, T., Longo, A., Murthy, M.N., Murthy, S., Dhavala, K., 2008. Counting the cost of vulture decline: an appraisal of the human health and other benefits of vultures in India. Ecol. Econ. 67 (2), 194–204. Martinez, G.P., 2011. Human hydatidosis disease: general background and epidemiological situation in Chile, 2001-2009. Rev. Chilena Infectol. 28 (6), 585–591. Mehta, R.B., Ananthkrishnan, N., Gupta, B.K., Srivastava, K.K., Mehdiratta, K.S., Satya, P., 1982. Hydatid disease in Pondicherry. Indian J. Surg. 44, 88–94. Moro, P.L., Budke, C.M., Schantz, P.M., Vasquez, J., Santivan, S.J., 2011. Economic impact of cystic echinococcosis in Peru. PLoS Negl. Trop. Dis. 5 (5), e1179.
12
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
NABARD, 2010. Model Bankable Projects – Animal Husbandry. National Bank for Agriculture and Rural Development, Government of India. Parija, S.C., Rao, R.S., Badrinath, S., Sengupta, D.N., 1983. Hydatid disease in Pondicherry. J. Trop. Med. Hyg. 86, 113–115. Polydorou, K., 1981. Animal health and economics. Case study: echinococcosis with a reference to Cyprus. Bull. Off. Int. Epizoot. 93, 981–992. Ranjan, S.K., Rawat, A.S., 2011. Indian Meat Industry Red Meat Manual (Annexure VI). Published by Agricultural & Processed Food Products Expert Development Authority, Government of India. Rao, S.S., Mehra, B., Narang, R., 2012. The spectrum of hydatid disease in rural central India: an 11-year experience. Ann. Trop. Med. Public Health 5, 225–230. Rausch, R.L., 1986. Life cycle patterns and geographic distribution of Echinococcus species. In: Thompson, R.C.A., Allen, G.A., Unwin. (Eds.), The Biology of Echinococcus and Hydatid Disease. George Allen & Unwin, London, pp. 44–80. Rausch, R.L., 1995. Life cycle patterns and geographic distribution of Echinococcus species. In: Thompson, R.C.A., Allen, G.A. (Eds.), Echinococcus and Hydatid Disease. CAB International, Wallingford, Oxon, pp. 88–134. Singh, B.B., 2011. Molecular Epidemiology of Echinococcosis in Northern India and its Public Health Significance, Ph. D Dissertation submitted to Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana, Punjab (India). Singh, B.B., Sharma, R., Gill, J.P.S., Aulakh, R.S., Banga, H.S., 2011. Climate change, Zoonoses and India. Rev. Sci. Tech. Off. Int. Epiz. 30 (3), 779–788. Singh, B.B., Sharma, R., Sharma, J.K., Juyal, P.D., 2010. Parasitic zoonoses in India – an overview. Rev. Sci. Tech. Off. Int. Epiz. 29 (3), 629–637. Singh, B.B., Ghatak, S., Banga, H.S., Gill, J.P.S., Singh, B., 2013. Veterinary urban hygiene – challenge for India. Rev. Sci. Tech. Off. Int. Epiz. 32 (3), in press.
Singh, G.S., Prabhakar, S., Ito, A., Cho, S.Y., Qiu, D., 2002. Taenia solium Taeniasis and Cysticercosis in Asia. In: Taenia solium Cysticercosis from Basic to Clinical Science. CABI Publishing, Oxon, UK, pp. 111–128. Thompson, R.C.A., 1995. Biology and systematics of Echinococcus. In: Thompson, R.C.A., Lymbery, A.J. (Eds.), Echinococcus and Hydatid Disease. CAB International, Wallingford, pp. 1–50. Thompson, R.C.A., Allsopp, C.E., 1988. Hydatidosis. Veterinary Perspectives and Annotated Bibliography. C.A.B. International, Wallingford, Oxon, pp. 246. Tiwary, A.K., Tiwary, R.N., 1988. Hydatid disease in Chotanagpur region of South Bihar. Indian J. Surg. 50, 14–18. Torgerson, P.R., Dowling, P.M., Abo-Shehada, M.N., 2001. Estimating the economic effects of cystic echinococcosis. Part 3: Jordan, a developing country with lower-middle income. Ann. Trop. Med. Parasitol. 95 (6), 595–603. Torgerson, P.R., 2003. Economic effects of echinococcosis. Acta Trop. 85 (2), 113–118. Torgerson, P.R., Carmona, C., Bonifacino, R., 2000. Estimating the economic effects of cystic echinococcosis: Uruguay, a developing country with upper-middle income. Ann. Trop. Med. Parasitol. 94, 703–713. Vamsy, M., Parija, S.C., Sibal, R.N., 1991. Abdominal hydatidosis in Pondicherry, India, Southeast Asian. J. Trop. Med. Public Health 22, 365–370. WHO, 1984. In: Eckert, J., Gemmell, M., Matyas, Z., Soulsby, E.J.L. (Eds.), Guidelines for Surveillance, Prevention and Control of Echinococcosis/Hydatidosis. World Health Organization, Geneva, pp. 1–147. Wright, I.A., Deka, R., Thorpe, W., Lapar, M.L., 2010. The pig sector in northeast India: status, constraints and opportunities. In: International Symposium on Sustainable Land Use and Rural Development in Mountainous Regions of Southeast Asia, Honoi.
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Economic losses due to cystic echinococcosis in India: Need for urgent action to control the disease Balbir B. Singh a,∗ , Navneet K. Dhand b , Sandeep Ghatak c , Jatinder P.S. Gill a a
School of Public Health & Zoonoses, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana, Punjab 141004, India Faculty of Veterinary Science, The University of Sydney, 425 Werombi Road, Camden, 2570 NSW, Australia c Division of Animal Health, ICAR Research Complex for the North Eastern Hill Region, Umroi Road, Umiam, Barapani, Meghalaya 793103, India b
a r t i c l e
i n f o
Article history: Received 31 March 2013 Received in revised form 22 July 2013 Accepted 14 September 2013 Keywords: Hydatidosis Echinococcosis Economic loss India Developing countries Parasitic zoonosis
a b s t r a c t Cystic ehinococcosis (CE) caused by Echinococcus granulosus remains a neglected zoonotic disease despite its considerable human and animal health concerns. This is the first systematic analysis of the livestock and human related economic losses due to cystic echinococcosis in India. Data about human cases were obtained from a tertiary hospital. Human hydatidosis cases with and without surgical interventions were extrapolated to be 5647 and 17 075 per year assuming a total human population of 1 210 193 422 in India. Data about prevalence of hydatid cysts in important food producing animals were obtained from previously published abattoir based epidemiological surveys that reported a prevalence of 5.39% in cattle, 4.36% in buffaloes, 3.09% in pigs, 2.23% in sheep and 0.41% in goats. Animal population data were sourced from the latest census conducted by the Department of Animal Husbandry, Dairying and Fisheries, India. Other input parameters were obtained from published scientific literature. Probability distributions were included for many input values to account for variability and uncertainty. Sensitivity analyses were conducted to evaluate the effect of important parameters on the estimated economic losses. The analysis revealed a total annual median loss of Rs. 11.47 billion (approx. US $ 212.35 million). Cattle and buffalo industry accounted for most of the losses: 93.05% and 88.88% of the animal and total losses, respectively. Human hydatidosis related losses were estimated to be Rs. 472.72 million (approx. US $ 8.75 million) but are likely to be an under-estimate due to under-reporting of the disease in the country. The human losses more than quadrupled to Rs. 1953 million i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased to 0.2% in the sensitivity analyses. The social loss and psychological distress were not taken into account for calculating human loss. The results highlight an urgent need for a science based policy to control and manage the disease in the country. © 2013 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding author. Tel.: +91 161 2414009; fax: +91 161 2400822. E-mail address: [email protected] (B.B. Singh). 0167-5877/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.prevetmed.2013.09.007
Cystic echinococcosis (CE) is an important zoonosis caused by Echinococcus granulosus affecting both the livestock and human populations, particularly in developing countries such as India (Singh et al., 2010). Dogs act as a definitive host of E. granulosus; humans and food
2
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
producing animals acquire the infection by ingestion of eggs shed by the definitive host (Thompson, 1995). Cysts of E. granulosus have been found in a variety of intermediate and aberrant hosts such as cattle, buffalo, sheep, goat, pigs, horses, camel and man (Rausch, 1986, 1995; Thompson and Allsopp, 1988; WHO, 1984). In livestock, CE may lead to reduced yield and quality of meat, milk and wool; reduced birth rate; delayed performance and growth; and condemnation of organs, especially of liver and lung (Benner et al., 2010; Budke et al., 2005; Torgerson et al., 2000; Polydorou, 1981). Economic losses also occur due to destruction of infected viscera and dead animals or due to ban on export of animals and their products if these are required to be free of CE. Human echinococcosis continues to be a significant public health problem in several regions of the world. Human losses primarily occur due to the need for surgical interventions, hospital care and productivity losses in asymptomatic cases. A recent Spanish study reported high economic losses at D 148 964 534 with human disease constituting most (89.1%) of the total losses (Benner et al., 2010). In Chile, human hydatidosis was associated with the average mortality rate of 0.2 deaths per 100 000 inhabitants per annum and a loss of 3349 years of life due to the premature death of 235 people (Martinez, 2011). The rural population of India is dependent on livestock for their livelihood (Singh et al., 2011) where cattle and buffalo production systems constitute the backbone of the livestock industry. Hydatid cysts have been found in most of the food producing animals in India such as cattle, buffalo, sheep, goat and pigs. Presence of stray animals such as dogs and cattle, unorganised slaughtering, free access of dogs to slaughter waste, decline in vulture population, sewage irrigation, and improper garbage disposal methods increase the chances of disease transmission to humans (Singh et al., 2002). Despite being a significant zoonotic problem, not many studies have been conducted in India to understand disease epidemiology or to estimate economic losses due to the disease. As a result, virtually no programs have been developed or implemented at the national level to control or eradicate the disease. This is the first systematic analysis of the economic losses due to CE in India. Such studies are required to help develop disease prevention and control strategies for India and other developing countries. 2. Methods 2.1. Livestock related losses The important food producing animals that make a substantial contribution to the meat and dairy industry in the country—sheep, goat, cattle, buffalo and pigs—were included in the analysis. The prevalence of CE in livestock was based on a recent abattoir based epidemiological survey conducted in India (Singh, 2011). The official data for animal populations (Table 1) were obtained from the latest census carried out by the Department of Animal Husbandry, Dairying and Fisheries, India (DAHD & F, 2010). The annual wool output per sheep was taken from published scientific literature (Banerjee, 1991) while the average milk
Table 1 Cost parameters used to estimate the economic losses associated with cystic echinococcosis in livestock, India. Average cost, in Rs. per kg
Reference
Sheep Sheep carcass
131.26
Sheep liver
131.26
Ranjan and Rawat (2011) Ranjan and Rawat (2011) Market value NABARD (2010)
Parameter
Sheep lung Sheep wool
40 38
Goats Goat carcass
131.26
Goat liver
131.26
Goat lung Goat’s milk at farm gate Cattle Beef carcass Cow liver Cow lung Cow’s milk Buffalo Buffalo carcass
40 20 112.18 60 35 26
Ranjan and Rawat (2011) Market value Market value Market value
Buffalo liver Buffalo lung Buffalo’s milk
60 35 30
Ranjan and Rawat (2011) Market value Market value Market value
Pigs Pig carcass Pig liver Pig lungs
95 50 30
Wright et al. (2010) Market value Market value
Energy (kW h)
112.18
Ranjan and Rawat (2011) Ranjan and Rawat (2011) Market value Market value
2.82
Punjab State Electricity Regulatory Commission (2010) Punjab (India)
yield and carcass weight were sourced from the official data (DAHD & F, 2010). The drought power production equivalence was taken from previous studies (Banerjee, 1991). The price of energy (kW h) charged by Punjab State Electricity Regulatory Commission (2010) Punjab (India) was used to calculate losses in draught power. The other related input parameters such as organ weight, animal produce, productivity losses (decrease in fecundity, carcass weight, milk production, draught power and wool output), life expectancy and reproductive rates were obtained from the published scientific literature (Table 2). The prices of animal carcasses, milk, liver and lungs were sourced through market surveillance or from published scientific literature (Table 1). Both direct and indirect economic losses for all the species were calculated separately. While direct losses included those occurring due to offal condemnation of infected animals, the indirect losses included those occurring due to reduction in growth and production and decrease in fecundity. All the analyses were conducted using R-statistical program (R-statistical package version 2.12.0, R Development Core Team, http://www.r-project.org).
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
3
Table 2 Epidemiological parameters used to estimate the economic losses associated with cystic echinococcosis in livestock, India, 2007. Parameter Sheep Population Total population Total breedable female population No. of sheep slaughtered per year Prevalence of infection At inspection In liver In lungs Production Sheep carcass Sheep liver Sheep lung Wool per sheep Mean lambing per year per ewe No. unborn lambs Lamb mortality rate Goats Population Total population Total breedable female In milk No. of goats slaughtered per year Prevalence of infection At inspection In livers In lungs Production Average carcass weight Goat liver Goat lung Mean kidding per year per doe Average milk yield of goat Lactation length No. unborn kids Kid mortality Cattle Population Total population Total breedable female population In milk No. of slaughtered cattle per year Prevalence of infection At inspection In livers In lungs Production Average carcass weight Cattle liver Cattle lung Mean calving per year per cow Average milk yield of cattle Drought power production equivalence Calf mortality No. of unborn calves Average working hrs per year per drought cattle Buffalo Population Total population In milk Breedable female population No. of slaughtered Buffalo per year Prevalence of infection At inspection In livers In lungs
Value
Range
Unit
Distribution
Reference
71 558 000 53 684 000 23 269 000
NA NA NA
Individual Individual Individual
Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
2.23 1.44 1.32
NA NA NA
% % %
Beta (18, 742) Beta (12, 748) Beta (11, 749)
Singh (2011) Singh (2011) Singh (2011)
13 566.5 623.5 1.0 1.25 82 304 10
453–680 340–907 NA 0.9–1.6 NA NA
kg g g kg/year Individual Individual %
Fixed Uniform Uniform Fixed Uniform Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) Banerjee (1991) Calculation NABARD (2010)
140 537 000 62 489 000 28 868 000 50 707 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
0.41 0.28 0.24
NA NA NA
% % %
Beta (11, 2428) Beta (8, 2431) Beta (7, 2432)
Singh (2011) Singh (2011) Singh (2011)
10 566.5 623.5 2.67 0.37 160 37 624 15
NA 453–680 340–907 2.01–3.33 NA NA NA NA
kg g g Individual kg/day day Individual %
Fixed Uniform Uniform Uniform Fixed Fixed Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Banerjee (1991) Calculation NABARD (2010)
199 075 000 72 915 000 38 928 000 2 476 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
5.39 4.68 4.32
NA NA NA
% % %
Beta (16, 262) Beta (14, 264) Beta (13, 265)
Singh (2011) Singh (2011) Singh (2011)
90 5.4 2.6 1 4.505 0.75
NA NA 2.2–3.0 NA 2.14–6.87 NA
kg kg kg Individual kg/day HP
Fixed Fixed Uniform Fixed Uniform Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Banerjee (1991)
5 216 157 360
NA NA NA
% Individual h
Fixed Fixed Fixed
NABARD (2010) Calculation Calculation
105 343 000 35 479 000 54 475 000 5 884 000
NA NA NA NA
Individual Individual Individual Individual
Fixed Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
4.36 4.36 4.36
NA NA NA
% % %
Beta (14, 284) Beta (14, 284) Beta (14, 284)
Singh (2011) Singh (2011) Singh (2011)
4
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Table 2 (Continued) Parameter Production Average carcass weight Buffalo liver Buffalo lung Mean calving per year Average milk yield of dairy buffalo No. of unborn calves Calf mortality Pigs Population Total population No. of pigs slaughtered Total no. of breedable female Prevalence of infection At inspection In livers In lungs Production Average weight of pig carcass Pig lung Pig liver Mean no. of piglets per year No. of unborn pigs Piglet mortality Productivity losses—all livestock Decrease in fecundity Decrease in carcass weight Decrease in milk production Decrease in drought power output Decrease in wool output Productive lifespan Farmer’s benefit from unborn animal produce
Value
Range
Unit
Distribution
Reference
106 5.4 2.6 1 4.57 130 631 5
NA NA 2.2–3.0 NA NA NA NA
kg kg kg Individual kg/day Individual %
Fixed Fixed Uniform Fixed Fixed Fixed Fixed
DAHD & F (2010) Gracey and Collins (1992) Gracey and Collins (1992) Banerjee (1991) DAHD & F (2010) Calculation NABARD (2010)
11 134 000 6 746 000 3 055 000
NA NA NA
Individual Individual Individual
Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010)
3.09 2.82 1.97
NA NA NA
% % %
Beta Beta Beta
Singh (2011) Singh (2011) Singh (2011)
35 396.5 1.45 16 83 072 20
NA 340–453 0.9–2 NA NA NA
kg g kg Individual Individual %
Fixed Uniform Uniform Fixed Fixed Fixed
DAHD & F (2010) DAHD & F (2010) DAHD & F (2010) Banerjee (1991) Calculation NABARD (2010)
5.5 6.25 2.5 2 2.5 35 15
0.0–11.0 2.5–10.0 0.0–5.0 NA NA NA 10–20
% decrease per year % decrease per year % decrease per year % decrease per year % decrease per year % of average lifespan %
Triangular Triangular Triangular Triangular Triangular Fixed Uniform
Benner et al. (2010) Benner et al. (2010) Benner et al. (2010) Benner et al. (2010) See text See text See text
2.1.1. Direct losses The direct losses occurring due to offal condemnation were calculated based on average weight and market price for liver and lung for all the five species taking into account the infected slaughtered animal population in India. For example, the direct losses due to condemned lungs for cattle were calculated as follows: Number of infected condemned lungs = Number of cattle slaughtered per year × Prevalence of infection in lungs
2.1.2.1. Losses due to reduction in milk production. To estimate losses due to reduction in milk yield in the cattle, buffalo and goat industry, we first calculated the number of infected ‘in milk’ animals during the year (Eq. (3)) and their lactation yield (Eq. (4)) from the official data provided by Department of Animal Husbandry, Dairying and Fisheries, India (DAHD & F, 2010). Annual milk losses were then calculated by taking into account the reduction in milk yield due to CE (Eqs. (5 and 6)) and the price of milk (Eq. (7)). The losses due to decrease in milk yield for the other two dairy species were calculated in a similar manner.
(1) Infected ‘in milk’ animals = Total number of cows in milk × Prevalence of infection
(3)
Losses due to condemned infected lungs = Number of lungs condemned × Average weight of lung × Price of lung
(2)
Milk produced per animal per lactation = Average milk production per animal per day
Similarly, the losses due to condemnation of liver and lungs were estimated for all the species. The total direct losses were calculated by combining losses from all species. 2.1.2. Indirect losses Decrease in milk production, growth and fecundity of infected animals were the important indirect losses considered in this study.
× Lactation length (days)
(4)
Milk loss per animal per annum due to CE = Milk produced per animal per lactation × Proportional reduction in milk production due to CE per year
(5)
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Total milk loss due to CE = Milk loss per animal due to CE × Infected in milk animals
(6)
Cost of lost milk = Total milk loss due to CE × Price of milk (7)
2.1.2.2. Losses due to reduction in carcass weight. To calculate losses due to reduced carcass weight at the time of slaughter, first the number of infected animals slaughtered during the year were calculated (Eq. (8)), followed by reduction in carcass weight per animal (Eqs. (9 and 10)). Annual losses due to reduced carcass weight were then estimated by including the price of meat (Eq. (11)). The Eqs. (8–11) show calculation of indirect losses due to reduced carcass weight for one species; losses for other species were also calculated in a similar manner. No. of infected animal slaughtered per year
5
To estimate economic losses, we estimated the numbers of males and females in these infected and uninfected groups but equations are only shown for the healthy offspring (Eq. (17)). Of the unborn male offspring that would have been slaughtered or used for draught purposes were calculated as shown in Eqs. (18–19). The losses from unborn male offspring without CE were then estimated for meat (Eqs. (20–21)) and for drought power (Eqs. (23–25)). Similarly, losses from female offspring were calculated as shown in Eqs. (26–28). No. of infected breedable females = Total breedable female population × Prevalence of infection
(12)
No. of unborn offspring = No. of infected breedable females × Mean numbers of offspring per year per animal × Decrease in fecundity
(13)
= No. of animal slaughtered per year × Prevalence of infection at inspection
(8)
No. of unborn offspring that would have survived = Unborn offspring
Loss of carcass weight per animal due to CE
× (1 − Newborn offspring mortality rate)
(14)
= Average weight of carcass × Percentage reduction in carcass weight
(9)
= Unborn offspring that would have survived
Total loss of carcass weight
× Prevalence of infection
= No. of infected animal slaughtered per year × Loss of carcass weight per animal due to CE
(10)
(15)
No. of healthy unborn offspring without CE = Unborn offspring that would have survived
Cost of lost carcass weight = Total loss of carcass weight × Price of meat
No. of unborn offspring infected with CE
(11)
2.1.2.3. Losses due to reduction in fecundity. Reduction in fecundity was calculated from infected breedable population (Eq. (12)) and the annual number of offspring (DAHD & F, 2010) as described by Benner et al. (2010). These calculations were based on the assumptions that prevalence of infection was equal in breeding and non-breeding populations. First we calculated the number of infected breedable females (Eq. (12)) and then the number of unborn offspring from them due to reduction in fecundity (Eq. (13)). However, not all of these unborn offspring would have survived because some of them would have died due to other reasons. This was accounted for in Eq. (14) and then the number of infected and uninfected unborn offspring that would have survived were estimated (Eqs. (15–16)) after assuming an equivalent prevalence of infection in unborn animals.
− Unborn offspring infected with CE
(16)
No. of unborn (male/female) offspring without CE = Healthy unborn offspring without CE × Percentage of (male/female) offspring born
(17)
No. of unborn male offspring without CE that would have been slaughtered = Unborn male offspring without CE × Proportion of male animals that undergo slaughter
(18)
Unborn male offspring without CE that would have been used for draught = Unborn male offspring without CE × Proportion of male animals used for draught purpose
(19)
6
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Losses due to decrease in fecundity
Productive lifespan = Average lifespan × Percentage of lifespan in production
(20)
= Losses from healthy unborn offspring without CE + Losses from healthy unborn offspring with CE
(30)
Meat loss due to CE = Healthy unborn male offspring without CE that would have been slaughtered × Average weight of carcass
Actual loss in farmers’ profit due to decrease in fecundity (21)
× Farmer’s profit percentage
Cost of lost meat = Meat loss due to CE × Price of meat (22) Draught energy produced per animal over lifetime = Av draught power equivalence per animal
(31)
The indirect losses due to decrease in fecundity for other species were calculated in similar fashion with slight modifications. For example, sheep wool losses were calculated as follows: Wool produced per animal over lifetime
× Working hours per year per animal × Productive lifespan
= Total loss due to decrease in fecundity
(23)
= Productive lifespan × Average wool produced per animal per year
(32)
Total loss of draught power Wool loss due to CE = Healthy unborn offspring without CE
= Draught energy produced per animal over lifetime × Healthy unborn male offspring without CE that would have been used for draught purposes
× Wool produced per animal over lifetime
(33)
(24) Cost of lost wool = Wool loss due to CE × Price of wool
Cost of lost draught power =
(34)
Total loss of draught power 1000
× Price of draught energy/kW h
(25)
Milk produced per animal over lifetime = Milk produced per animal per lactation × Productive lifespan
(26)
Total milk loss from unborn healthy female offspring = Milk produced per animal over lifetime × Healthy unborn female offspring without CE
(27)
Cost of lost milk = Total milk loss from unborn healthy female offspring × Price of milk
(28)
Total losses were calculated by combining losses from meat, drought power and milk (Eq. (29)). Losses from unborn offspring infected with CE were also calculated in the similar manner. The total losses due to decrease in fecundity were calculated as shown in Eqs. (30 and 31). Losses from healthy unborn offspring without CE = Cost of lost meat + Cost of lost drought power + Cost of lost milk
(29)
Finally, overall losses were calculated from decreased fecundity, meat and wool losses in sheep; meat and milk losses in goat; meat, milk and draught power losses in cattle; milk and meat losses in buffalo and meat losses in pigs. The buffalo industry in India does not have any distinct meat and dairy breeds. In this industry, females are reared for milk and males for meat purposes. Therefore, the calculations were done assuming equal populations of unborn male and female animals. The scenario of the cattle industry in the country is different due to the ban on cow slaughtering in most parts of the country. The female animals are reared for milk but the fate of male animals is not fully known. Therefore, to calculate annual losses due to decrease fecundity in cattle industry, we assumed that all of the female population was reared for milk, 5% of the male population was slaughtered and 30% of the male population was used for draught power due to partial mechanization of Indian agriculture. As per the official estimates, slaughtered cattle population were approximately 3.5% of the total male populations. However, to adjust for unaccounted or illegal slaughtering practices, we assumed that 5% of the male population was slaughtered. Similarly, the male population used for other purposes except for breeding was 60% of the total male population. We assumed that at least half of that population (30%) is used for drought purposes. As cattle bulls are used in agricultural operations in India, the decrease in draught power for draught cattle was also estimated. After calculating lifetime costs of animal produce and draught power from unborn animals, we assumed that
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
7
Table 3 Epidemiological and economic parameters used to estimate the human economic losses associated with cystic echinococcosis in India. Parameter
Value
Distribution
Reference
Total number of diagnosed cases without surgery Total number of diagnosed cases with surgical/interventional procedure Undiagnosed or asymptomatic cases Productivity loss as % per year Cost of diagnosed cases without surgery Cost of diagnosed cases involving surgical/interventional procedures Per capita income (per annum)*
17 075 5646
Fixed Fixed
Khurana et al. (2007) Khurana et al. (2007)
0.00–0.021–0.04 0–4 Rs. 2000 Rs. 50 000
Triangular Uniform Fixed Fixed
See text Benner et al. (2010) Personnel communication Personnel communication
Rs 33 283
Fixed
1 210 193 422
Fixed
Ministry of Statistics and Programme Implementation, Govt. of India Census of India, 2011
Total population *
For the year 2007–2008.
farmer would receive 15% (range 10–20%) benefits from rearing of these animals after subtracting all the other investment costs. This assumption was used to calculate actual loss in farmer profits due to decrease in fecundity. 2.2. Human related losses 2.2.1. Sources of data The human epidemiological parameters used in the analysis included the prevalence of diagnosed cases with surgical intervention, diagnosed cases without surgical intervention and undiagnosed or asymptomatic cases. To estimate the productivity losses in humans, the number of reported CE cases was obtained from analysis of the retrospective data of a tertiary hospital (Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India) (Khurana et al., 2007). On the average, 1 318 271 patients visited the hospital per year during 2007–2009 (42nd annual report, PGIMER). During the past 20 years 495 cases were diagnosed with CE of which 123 required surgical intervention and 372 were treated without surgical intervention. The total numbers of diagnosed cases in the country were extrapolated from the PGIMER data (Table 3). 2.2.2. Estimation of costs for treating cases The fixed costs were calculated for two types of cases: treatment of diagnosed cases who had gone for surgery to remove cysts (Eq. (35)) and treatment of patients who were diagnosed but had not gone for removal of cysts (Eq. (36)).
2.2.3. Estimation of costs due to loss of productivity Losses due to productivity were estimated for three types of cases: (a) diagnosed cases requiring surgery; (b) diagnosed cases without surgery; and (c) undiagnosed cases. To calculate productivity losses, official per capita income per annum was obtained and productivity losses due to CE were taken from published scientific literature (Eq. (37)). The productivity losses associated with diagnosed cases with and without surgical intervention were estimated as shown in Eqs. (38–39). Loss in annual per capita income due to CE = Annual per capita income × Annual productivity percentage loss
(37)
Loss in annual per capita income due to CE from diagnosed cases without surgery = Loss in annual per capita income due to CE × Total number of diagnosed cases without surgery
(38)
Loss in annual per capita income due to CE from diagnosed cases with surgical or interventional procedure = Loss in annual per capita income due to CE × Total number of diagnosed cases with surgical or
Losses due to treating CE diagnosed cases with surgery
interventional procedure
(39)
= Total number of diagnosed cases with surgical or interventional procedure × Cost of treating diagnosed cases involving surgical or interventional procedures
(35)
Losses due to treating CE diagnosed cases without surgery = Total number of diagnosed cases without surgery × Cost of treating diagnosed cases without surgery
(36)
The productivity losses associated with undiagnosed cases were not possible to be directly estimated because undiagnosed or asymptomatic cases have never been identified in India using ultrasound. Therefore, the prevalence of these cases was calculated based on a similar study carried out in Florida (Uruguay), where prevalence of undiagnosed or asymptomatic cases was found to be 1.64% by ultrasound, with annual surgical incidence of 36.1 cases per 105 person-years, and a ratio of 45.4 (Benner et al., 2010; Carmona et al., 1998). First, incidence rate of surgical cases in India was calculated to be 0.466 cases per
8
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
105 person-years based on the total cases requiring surgical intervention (5646). From this estimate, the mean prevalence of undiagnosed or asymptomatic CE cases was calculated as 0.021%. These calculations were based on comparison between annual surgical incidence in India with prevalence of undiagnosed or asymptomatic cases by ultrasound and annual surgical incidence in Florida (Uruguay).
Mean prevalence of undiagnosed or asymptomatic CE cases in India
=
Prevalence of undiagnosed or asymptomatic cases in Florida
Annual surgical incidence per 105 person years in Florida
× Annual surgical incidence per 105 person years in India
(40)
2.4. Sensitivity analyses Sensitivity analyses were conducted to evaluate the effect of changing input values of prevalence of undiagnosed asymptomatic cases and per capita income (per annum) for which objective information was not available. Budke et al. (2006), assumed that ≈10% of annual cases are not officially diagnosed due to socioeconomic status of the patients or the subclinical nature of the illness. Therefore sensitivity analysis was conducted to make adjustments for under-reporting. We assumed a very low proportion of asymptomatic cases; therefore, it was increased substantially in the sensitivity analysis. Similarly, we changed per capita income to evaluate the effect of income on losses due to the disease. 3. Results
Total annual loss in per capita income due to CE from undiagnosed cases = Loss in per capita income per year due to CE × Total number of undiagnosed cases
(41)
2.3. Accounting for variability and uncertainty
In total, CE was found to cause median economic losses to the amount of Rs. 11.47 billions (approx. US $ 212.35 million) (95% CI 6.63–19.48 billion) per annum (Table 4). Median livestock and human related losses amounted to Rs. 10.95 billion (95% CI 6.11–18.74 billion) and 0.472 (95% CI 0.330–0.737 billion), respectively, comprising 95.52% and 4.12% of the total losses (Fig. 1). 3.1. Livestock related losses
Beta probability distributions were used to account for uncertainties in the prevalence of CE. Alpha (˛) and beta (ˇ) parameters of beta probability distributions were based on the recent abattoir based epidemiological survey conducted in India (Singh, 2011).
The economic losses related to cattle, buffalo, sheep, goat and pigs are provided in Table 4. Occurrence of CE in cattle and buffalo accounted for 93.05% of the total livestock and 88.88% of the total (livestock and human) losses.
˛ = number of positive CE cases + 1.
3.2. Human related losses
ˇ = total number of cases examined − number of positive CE cases + 1. Uniform distributions were used for weight of the condemned offal. The minimum and maximum values for weight of the condemned liver and lungs were taken from the published scientific literature (Gracey and Collins, 1992). Triangular distributions were applied for decrease in fecundity, carcass weight, milk production, draught power, wool output. For calculating actual farmer profits due to decrease in fecundity, we applied uniform distribution in the range of 10–20% because benefits are likely to vary under different rearing conditions. We used a triangular distribution with a mode of 0.021% and a range of 0.00–0.04 for the proportion of undiagnosed or asymptomatic human cases for analysis of economic losses. Uniform distributions were applied to calculate productivity loss (0.0–4.0%) in diagnosed and undiagnosed cases (Torgerson et al., 2000). The similar distributions were also applied in a Spanish study (Benner et al., 2010). The confidence limits for the estimates were calculated and running Monte Carlo simulations for 10 000 iterations.
CE was found to cause median human losses amounting Rs. 472.7 million (approx. US $ 8.75 million). 3.3. Sensitivity analysis The sensitivity analysis revealed that annual median losses were Rs. 496.9 million (95% CI 334–652 million) (approx. US $ 9.20 million) at the human undiagnosed prevalence of 0.02 (Fig. 2). However, the annual median losses increased to Rs. 1953 million (95% CI 488–3413 million), i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased 10 times to 0.2%. The sensitivity analysis further revealed that human losses were Rs. 362.9 million (95% CI 320.4–442 million), i.e. approx. US $ 6.72 million, assuming the annual per capita income of Rs. 10 000 (Fig. 2). However the losses rose to Rs. 781.6 million (95% CI 358.2–1578 million), i.e. approx. US $ 14.47 million when the annual per capita income of Rs. 100 000 was assumed. 4. Discussion As far as we are aware, this is the first systematic analysis of economic losses occurring due to CE in India. The results indicate that CE causes huge economic losses in India at the estimated prevalence rates of the
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
9
Table 4 Total losses associated with cystic echinococcosis in humans and livestock in India estimated in the study. Total losses
Indian rupees Median
Sheep Goat Cattle Buffalo Pig Human Total animal loss Grand total loss
113 787 069 53 274 837 5 599 822 658 4 592 145 524 96 141 417 472 715 931 10 952 980 207 11 467 000 000
US dollars 95% confidence intervals 75 875 331–167 202 569 32 850 991–84 511 258 2 133 969 312–12 588 894 415 2 071 978 664–9 275 390 197 55 561 238–1 59 243 769 330 337 589–737 059 556 6 105 879 161–18 736 319 358 6 629 764 108–19 478 312 127
Median 2 107 168 986 571 103 700 420 85 039 732 1 780 397 8 753 999 202 832 967 212 351 852
95% confidence intervals 1 405 099–3 096 344 608 352–1 565 023 39 517 950–233 127 674 38 369 975–171 766 485 1 028 912–2 948 959 6 117 363–13 649 251 113 071 836–346 968 877 122 773 409–360 709 484
Fig. 1. Total, livestock and human associated losses estimated in the study due to cystic echinococcosis in India.
10
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Human loss (Million rupees)
a) Human undiagnosed cases 4000 3500 3000 2500 2000 1500 1000 500 0
0.02
0.05
0.1
0.15
0.2
Human loss (Million rupees)
Proporon of undiagnosed cases b) Per capita income
1800 1600 1400 1200 1000 800 600 400 200 0
10
20
30
40
50
60
70
80
90
100
Per capita income (thousand rupees) Fig. 2. Sensitivity analysis of a) asymptomatic human undiagnosed cases and b) annual per capita income to account for losses associated with human cystic echinococcosis. The tick marks indicate the median losses and the bars extend to the 95% lower and upper confidence intervals.
infection. Globally, Budke et al. (2006) estimated human annual loss of US $ 194 million (US $ 171–217 million). When they accounted under-reporting, the losses increased up to US $ 763 million (US $ 676–857 million). An annual livestock production loss of at least US $ 141 million (US $ 101–183 million) and possibly up to US $ 2190 million (US $ 1572–2951 million) was also estimated across the globe (Budke et al., 2006). High costs were estimated in several other countries also (Torgerson et al., 2001; Moro et al., 2011; Benner et al., 2010; Fasihi et al., 2012). All these results indicate that CE causes substantial economic losses across the globe and India is no exception. Estimation of economic losses of a zoonosis gives best indication of the combined human and livestock sector losses (Carabin et al., 2005). From the livestock sector, cattle and buffalo industries contributed most of the losses due to CE. Decrease in fecundity in cattle and buffalo represented the biggest indirect loss associated with CE. These losses could have serious consequences as dairy industry is the largest and most important livestock industry in India. However, the losses due to slaughter of infected cattle were fairly low. This disparity is due to religious and cultural beliefs in India. Cows are considered to be holy animals in India, particularly among Hindus, which constitute about 80% of the Indian population. Therefore, cow slaughter is banned in most of the Indian states (Singh et al., 2013). The cattle are mainly reared for milk production and upon death, the carcasses are usually left for scavenging. The figure of 2.47 million for the number of cattle slaughtered per year reflects only the cattle that are slaughtered each year. Occurrence of major losses in the livestock industry indicates that a greater
emphasis should be placed on controlling CE in animals particularly in cattle and buffalo. While indirect losses were a major component of the livestock related losses, the direct losses were significant for human CE associated economic losses. The human losses were influenced by prevalence of undiagnosed cases and more than quadrupled to Rs. 1953 million i.e. approx. US $ 36.17 million, when the prevalence of human undiagnosed cases was increased to 0.2% in the sensitivity analyses. Similarly, per capita income was also found to have substantial impact on human related economic losses. Additionally, CE causes huge psychological and socioeconomic losses which could not be accounted in the present study. Such losses could arise due to permanently decreased quality of life and a loss of job (Torgerson, 2003). We acknowledge that the human related prevalence rates of infection are likely to be an under-estimate because not everyone has access to advanced medical care facilities in the country, leading to under-reporting of the human diseases. Secondly, the asymptomatic or undiagnosed cases are likely to be an under-estimate as ultrasonography based surveys have not been conducted in the country and the prevalence of asymptomatic or undiagnosed cases was calculated from the studies carried out in other countries. Therefore, human related cost calculations are lower compared to economic losses due to human CE as reported across the globe (Budke et al., 2006) and in many other countries such as Spain (Benner et al., 2010), Peru (Moro et al., 2011), Jordon (Torgerson et al., 2001) and Iran (Fasihi et al., 2012). The disease control programmes usually rely on public education, upgrading of meat inspection facilities, dog population control and active detection and elimination of E. granulosus in infected dogs. However, presence of many factors such as huge stray dog population, unorganised slaughtering, free access of dogs to slaughter waste (Singh et al., 2013), decline in vulture population (Markandya et al., 2008) and routine contamination of pastures with dog faeces make it really difficult to control the disease in India. Holistic implementation of animal birth control programme at the country level to control the ever increasing stray dog populations is a must for India and many other developing countries so as to break the transmission cycle of the disease. Many of the costs assumed in our analysis for medical care were consensual fixed costs but are in fact likely to vary in different parts of the country. For example, we used estimates of human prevalence from a hospital in northern India and extrapolated it to the whole population of India. These input parameters could be biased but the detailed epidemiological studies on human hydatidosis have not been conducted in southern India. Many case reports of human hydatidosis from central and southern India (Rao et al., 2012; Akther et al., 2011; Tiwary and Tiwary, 1988; Mehta et al., 1982; Parija et al., 1983; Vamsy et al., 1991) indicate that the disease is endemic in these parts of the country as in northern India. Due to similar socioeconomic, hygienic and animal husbandry management practices followed throughout the country, we expect a similar prevalence in these parts of the country.
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
Similarly, many of the costs assumed for animal produce were also fixed. Although, we used different prices of meat to calculate losses due to infected lungs and liver but the same prices were used to calculate losses due to rest of the carcasses for all the animals. This was realistic because most of the slaughterhouses in the country sell carcasses at gross prices. We believe that these limitations will not cause much difference to the total calculated losses but a small over- or under-estimate could not be ruled out. The study had a number of strengths. To obtain valid estimates of economic losses due to animals and humans, we used official figures for most of the parameters. For the input parameters where official data were unavailable, data from previously published research were used. For many input parameters, we used probability distributions already used in the previous studies to account of variability and uncertainty (Benner et al., 2010; Budke et al., 2006; Torgerson et al., 2000, 2001; Heath et al., 2003; Baitursinov et al., 2004). However, we could not consider many potentially influential factors due to unavailability of data. For example, we used overall prevalence of infection for calculating production losses. The size and number of the cysts were not taken into account for calculating such losses because this information was not available. Similarly, the human losses were estimated based on national annual per capita income. It has certain disadvantages as it does not take into account informal labour such as family care giving. Moreover, the prevalence of infection could vary in people belonging to different economic classes and could lead to an over or under-estimate of economic losses. However, we applied triangular distributions to account for uncertainty in losses arising due to decrease in fecundity, carcass weight, milk yield, wool output and drought power. Moreover, we conducted sensitivity analyses to evaluate the influence of input parameters such as prevalence of undiagnosed asymptomatic cases and per capita income (per annum) for which objective information was not available. Similarly, other factors which could influence indirect livestock losses such as animal breed, diet and management conditions could not be taken into consideration. We also did not estimate decrease in hide value due to animals infected with CE. We also did not include partial compensation through replacement for losses due to unborn animals, as most of the farmers keep only a few animals and do not have enough resources to replace for their unborn animals. Within the constraints of the study, we believe that livestock and human CE losses calculated in the present study represent the human and livestock related losses occurring due to CE in India. However, further studies should be conducted to refine the models and obtain precise information about input parameters from various parts of the country. 5. Conclusion CE is an important zoonosis affecting most of the food producing animals and man in India. The economic analysis revealed heavy losses associated with CE in the dairy industry. High prevalence of the disease in the cattle and buffalo were found to be important factors for huge economic losses. Human losses are likely to be an under-estimate due
11
to under-reporting of the disease in the country. There is urgent need for action based on implementation of intervention policies such as control of stray animal populations, prophylactic anthelmintic treatment of dogs, organised slaughtering and use of proper garbage disposal systems. Acknowledgements This study was financially supported by the School of Public Health & Zoonoses, Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana, Punjab, India. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j. prevetmed.2013.09.007. References Akther, M.J., Khanam, N., Rao, S., 2011. Clinico epidemiological profile of hydatid diseases in Central India, a retrospective and prospective study. Int. J. Biol. Med. Res. 2 (3), 603–606. Banerjee, G.C. (Ed.), 1991. A Text Book of Animal Husbandry. , Seventh ed. Oxford & IBH Publishing Pvt. Ltd, Kolkata. Baitursinov, K.K., Shaikenov, B., Abdybekova, A.M., 2004. Distribution of Echinococcus amongst agricultural animals in Kazakhstan. In: Torgeson, P.R., Shaikenov, B. (Eds.), Echinococcosis in Central Asia: Problems and Solutions. Dauir Publishing House, Almaty, Kazakhstan, pp. 101–118. Benner, C., Carabin, H., Serrano, L.P.S., Budke, C.M., Carmena, D., 2010. Analysis of the economic impact of cystic echinococcosis in Spain. Bull. World Health Organ. 88, 49–57. Budke, C.M., Deplazes, P., Torgerson, P.R., 2006. Global socioeconomic impact of cystic echinococcosis. Emerg. Infect Dis. 12 (2), 296–303. Budke, C.M., Jiamin, Q., Qian, W., Torgerson, P.R., 2005. Economic effects of echinococcosis in a disease endemic region of the Tibetan Plateau. Am. J. Trop. Med. Hyg. 73 (1), 2–10. Carabin, H., Budke, C.M., Cowan, L.D., Willingham, A.L., Torgerson, P.R., 2005. Methods for assessing the burden of parasitic zoonoses: echinococcosis and cysticercosis. Trends Parasitol. 21, 327–333. Carmona, C., Perdomo, R., Carbo, A., Alvarez, C., Monti, J., Grauert, R., 1998. Risk factors associated with human cystic echinococcosis in Florida Uruguay: results of a mass screening study using ultrasound and serology. Am. J. Trop. Med. Hyg. 58, 599–605. DAHD & F, 2010. Basic Animal Husbandry Statistics. Department of Animal Husbandry, Dairying & Fisheries, Ministry of Agriculture, Government of India. Fasihi, H.M., Budke, C.M., Rostami, S., 2012. The monetary burden of cystic echinococcosis in Iran. PLoS Negl. Trop. Dis. 6 (11), e1915. Gracey, J.F., Collins, D.S., 1992. Meat Hygiene, Ninth ed. Bailliere Tindall, England. Heath, D.D., Jensen, O., Lightowlers, M.W., 2003. Progress in control of hydatidosis using vaccination—a review of formulation and delivery of the vaccine and recommendations for practical use in control programmes. Acta Trop. 85, 133–143. Khurana, S., Das, A., Malla, N., 2007. Increasing trends in seroprevalence of human hydatidosis in North India: a hospital based study. Trop. Doct. 37, 100–102. Markandya, A., Taylor, T., Longo, A., Murthy, M.N., Murthy, S., Dhavala, K., 2008. Counting the cost of vulture decline: an appraisal of the human health and other benefits of vultures in India. Ecol. Econ. 67 (2), 194–204. Martinez, G.P., 2011. Human hydatidosis disease: general background and epidemiological situation in Chile, 2001-2009. Rev. Chilena Infectol. 28 (6), 585–591. Mehta, R.B., Ananthkrishnan, N., Gupta, B.K., Srivastava, K.K., Mehdiratta, K.S., Satya, P., 1982. Hydatid disease in Pondicherry. Indian J. Surg. 44, 88–94. Moro, P.L., Budke, C.M., Schantz, P.M., Vasquez, J., Santivan, S.J., 2011. Economic impact of cystic echinococcosis in Peru. PLoS Negl. Trop. Dis. 5 (5), e1179.
12
B.B. Singh et al. / Preventive Veterinary Medicine 113 (2014) 1–12
NABARD, 2010. Model Bankable Projects – Animal Husbandry. National Bank for Agriculture and Rural Development, Government of India. Parija, S.C., Rao, R.S., Badrinath, S., Sengupta, D.N., 1983. Hydatid disease in Pondicherry. J. Trop. Med. Hyg. 86, 113–115. Polydorou, K., 1981. Animal health and economics. Case study: echinococcosis with a reference to Cyprus. Bull. Off. Int. Epizoot. 93, 981–992. Ranjan, S.K., Rawat, A.S., 2011. Indian Meat Industry Red Meat Manual (Annexure VI). Published by Agricultural & Processed Food Products Expert Development Authority, Government of India. Rao, S.S., Mehra, B., Narang, R., 2012. The spectrum of hydatid disease in rural central India: an 11-year experience. Ann. Trop. Med. Public Health 5, 225–230. Rausch, R.L., 1986. Life cycle patterns and geographic distribution of Echinococcus species. In: Thompson, R.C.A., Allen, G.A., Unwin. (Eds.), The Biology of Echinococcus and Hydatid Disease. George Allen & Unwin, London, pp. 44–80. Rausch, R.L., 1995. Life cycle patterns and geographic distribution of Echinococcus species. In: Thompson, R.C.A., Allen, G.A. (Eds.), Echinococcus and Hydatid Disease. CAB International, Wallingford, Oxon, pp. 88–134. Singh, B.B., 2011. Molecular Epidemiology of Echinococcosis in Northern India and its Public Health Significance, Ph. D Dissertation submitted to Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana, Punjab (India). Singh, B.B., Sharma, R., Gill, J.P.S., Aulakh, R.S., Banga, H.S., 2011. Climate change, Zoonoses and India. Rev. Sci. Tech. Off. Int. Epiz. 30 (3), 779–788. Singh, B.B., Sharma, R., Sharma, J.K., Juyal, P.D., 2010. Parasitic zoonoses in India – an overview. Rev. Sci. Tech. Off. Int. Epiz. 29 (3), 629–637. Singh, B.B., Ghatak, S., Banga, H.S., Gill, J.P.S., Singh, B., 2013. Veterinary urban hygiene – challenge for India. Rev. Sci. Tech. Off. Int. Epiz. 32 (3), in press.
Singh, G.S., Prabhakar, S., Ito, A., Cho, S.Y., Qiu, D., 2002. Taenia solium Taeniasis and Cysticercosis in Asia. In: Taenia solium Cysticercosis from Basic to Clinical Science. CABI Publishing, Oxon, UK, pp. 111–128. Thompson, R.C.A., 1995. Biology and systematics of Echinococcus. In: Thompson, R.C.A., Lymbery, A.J. (Eds.), Echinococcus and Hydatid Disease. CAB International, Wallingford, pp. 1–50. Thompson, R.C.A., Allsopp, C.E., 1988. Hydatidosis. Veterinary Perspectives and Annotated Bibliography. C.A.B. International, Wallingford, Oxon, pp. 246. Tiwary, A.K., Tiwary, R.N., 1988. Hydatid disease in Chotanagpur region of South Bihar. Indian J. Surg. 50, 14–18. Torgerson, P.R., Dowling, P.M., Abo-Shehada, M.N., 2001. Estimating the economic effects of cystic echinococcosis. Part 3: Jordan, a developing country with lower-middle income. Ann. Trop. Med. Parasitol. 95 (6), 595–603. Torgerson, P.R., 2003. Economic effects of echinococcosis. Acta Trop. 85 (2), 113–118. Torgerson, P.R., Carmona, C., Bonifacino, R., 2000. Estimating the economic effects of cystic echinococcosis: Uruguay, a developing country with upper-middle income. Ann. Trop. Med. Parasitol. 94, 703–713. Vamsy, M., Parija, S.C., Sibal, R.N., 1991. Abdominal hydatidosis in Pondicherry, India, Southeast Asian. J. Trop. Med. Public Health 22, 365–370. WHO, 1984. In: Eckert, J., Gemmell, M., Matyas, Z., Soulsby, E.J.L. (Eds.), Guidelines for Surveillance, Prevention and Control of Echinococcosis/Hydatidosis. World Health Organization, Geneva, pp. 1–147. Wright, I.A., Deka, R., Thorpe, W., Lapar, M.L., 2010. The pig sector in northeast India: status, constraints and opportunities. In: International Symposium on Sustainable Land Use and Rural Development in Mountainous Regions of Southeast Asia, Honoi.
