Mortality rates adjusted for unobserved deaths and associations with Newcastle disease virus serology among unvaccinated village chickens in Myanmar

Abstract

Village chickens are an important livestock for many rural families in Myanmar and other developing countries. Village chickens are kept under free-ranging conditions, with confinement only at night. Therefore, it is likely that some deaths are not observed by farmers. We conducted a longitudinal study from November 2003 until May 2004 to describe temporal patterns of mortality of village chickens in 10 villages in Myanmar. Field veterinarians first identified the numbers of birds in all chicken-owning households in each village. We then selected 307 households randomly with stratification by flock size. Each study household was then visited once monthly at which time questionnaires were completed recording current flock structure and numbers of hatchings, mortalities, sales and birds consumed since the previous visit. In addition, sera were collected from a sample of adult birds and growers. Depending on month and age group of chicken, from 71 to 231 (out of 290–307) households had discrepancies in the counts of birds. For chicks, at least one-quarter of the households had unobserved losses of at least 5 chicks per household (maximum 66 chicks); unobserved losses were less for growers and adult chickens. The median month-specific, village-specific mortality rates per 1000 bird-days at risk (counting missing birds as deaths) ranged from 0.8 to 1.7 for adults, from 0.4 to 4.7 for growers and from 8.0 to 16.5 for chicks. Across all birds, the prevalence of protective titres against Newcastle disease virus was 79% (95% confidence interval 74, 84); higher prevalences of protective titres were associated with reduced mortality rates in the following months.