Graphical approaches to support the analysis of linear-multilevel models of lamb pre-weaning growth in Kolda (Senegal)

Evaluation of crude annual parturition rate estimates in a small-holder African ruminant farming system

Many parameters have been proposed for evaluating livestock reproduction performances in tropical farming systems. In tropical free-ranged and small-holder systems, where reproduction cycles cannot be individually observed without expensive field surveys, one of these parameters is the average number of parturitions (h) expected by reproductive female, if the female spends the whole year in the herd. A frequent approach for estimating h is to use the ratio hc = m/T, so-called 'crude annual parturition rate', where m is the observed number of parturitions and T is the total time of presence of the reproductive females in the herd during the year. The bias encountered when h is estimated by hc was evaluated in this paper. Six methods of estimation were used, where T was the exact observed time of presence (hc1) or approximated by monthly, quarterly, half-yearly and yearly averages or final size of the reproductive herd size (hc2 to hc6). Data came from long-term follow-up of cattle and small-ruminant herds (with data recorded at animal level) in extensive agro-pastoral systems in Senegal. In general, h was correctly estimated by hc1. Nevertheless hc1 was sensitive to competing risks (e.g. deaths, sales and slaughtering of reproductive females) and was seriously biased when intensive withdrawals of females occurred before or during the parturition peak. Reliability of crude rates progressively decreased from hc2 to hc6, corresponding to the degradation of information used for approximating T. This decrease was much lower for cattle (for which all methods had acceptable reliability) than for small ruminants. Among the compared methods, the lower reliability was observed for hc6 that we do not recommend for small ruminants. Methods hc5 and hc6 are currently used in rapid cross-sectional retrospective surveys based on the recall of the farmers on the demographic events which occurred in the herd over the last past 12 months. The study has showed that such surveys and estimates hc5 and hc6 can generate seriously biased results. More globally, annual parturition rates can be highly variable depending on the 12-month periods considered. Annual parturition rates estimated on short-term data, even with precise herd follow-up surveys, must be considered cautiously.

Assessment of age at first parturition by accounting censored data: the example of small ruminants in agropastoral herds in Senegal

Age at first parturition (AFP) is an important reproductive trait when evaluating the herd productivity in tropical small-holder farming systems, particularly for small ruminants with high demographic turnover and commercial utilization. The article reports the AFP probability distributions calculated on goats and sheep herds monitored for more than 10 years in three low-input livestock farming systems located from northern to southern Senegal (Louga, Kaymor, and Kolda) and representing different agroecological conditions and animal phenotypes. The AFP distributions were estimated with the Kaplan-Meier (KM) survival analysis accounting for censoring (females lost to follow-up before their first parturition). When measured by the KM estimator of the AFP median, the precocity for both species was almost 4 months later in Louga than in Kaymor and Kolda. In Louga, the precocity was 1.6 months later for sheep than for goats and 2.9 months in Kaymor and Kolda. These estimates were higher than those obtained with the common and crude approach that removes censored data before calculation (relative bias ranged from -12% to -22% for the AFP mean depending on species and site) and those already published in previous studies on the same data sets. In on-farm small ruminant studies focusing on reproductive traits such as AFP distributions, we recommend using survival models (e.g., the KM method) to avoid potential biases due to data censoring.

Evaluation of 12-month interval methods for estimating animal-times at risk in a traditional African livestock farming system

Demographic parameters are useful for assessing productivity and dynamics of tropical livestock populations. Common parameters are the annual instantaneous hazard rates, which can be estimated by m/T (where m represents the number of the considered demographic events occurred during the year and T the cumulated animal-time at risk). Different approaches are encountered in the literature for computing T from on-farm survey data. One crude approach ("the 12-month interval approach") only uses estimations of herds' sizes at beginning and end of the year and aggregated counts of demographic events over the year. I evaluated the potential biases in using four 12-month interval methods (M1-M4) to estimate T. Biases were evaluated by comparing the 12-month estimates to gold-standard values of T. Data came from long-term herd monitoring on cattle and small ruminants in extensive agro-pastoral systems. Animal-times at risks were correctly estimated in average by methods M1, M2 and M4 (average relative biases<or=6% in absolute values), except for adult-male small ruminants. For young animals, M2 and M4 showed equivalent biases. M2 is simple to implement and has the advantage of being applicable for any age-group, although M4 is only applicable for young animals. M3 was highly biased and I do not recommend it. Although accurate in average, 12-month interval methods showed highly variable biases. This variability results from interactions between the dates delimiting the 12-month interval and the distributions of the demographic events over time. This phenomenon is particularly important for the adult-male small ruminants. Based on the bias variability observed in the study, the user of 12-month interval methods has to remember that they only provide approximate results and that they cannot completely replace the gold-standard approaches.

Use of Akaike information criteria for model selection and inference. An application to assess prevention of gastrointestinal parasitism and respiratory mortality of Guinean goats in Kolda, Senegal

A field experiment was carried out in Kolda (southern Senegal) from July 1986 to July 1988. Its goals were to: (1) describe the patterns of mortality of female Guinean goats by age, season and year; (2) assess preventive measures against respiratory diseases and gastrointestinal parasitism in reducing mortality; and (3) estimate the overall impact of these measures on survival to 1 year of age. Preventive measures for respiratory disease included vaccination against peste des petits ruminants (PPR) and pneumonic pasteurellosis (Pasteurella multocida types A and D). Control of gastrointestinal parasites was by deworming does with morantel (7.5mg kg(-1), three times during the rainy season). The effects of vaccines and deworming were tested in a randomised factorial field experiment with villages being the experimental units. A total of 19 villages, 113 goat herds and 1,458 goats were included in the study. Generalised linear models of survival for five cohorts of goats (defined by five different birth seasons) used a binomial assumption for the response distribution and a complementary log-log link. Explanatory variables included age, season, year, vaccination, deworming and their interactions. A complex a priori model was built on the basis of previous epidemiological knowledge; a purposely selected set of simpler models was compared to this full model by the Akaike information criterion (AIC) and derived statistics. Inference on 1-year survival and treatment effects accounted for model-selection uncertainty. It was carried out with a bootstrap procedure and used information from the whole set of selected models. Large variations in mortality by year and season were observed but no regular seasonal pattern was apparent. Mortality probabilities of kids in dewormed groups decreased quickly after birth, but remained elevated up to 9 months of age in the non-dewormed groups. Deworming lowered the risk of mortality. Vaccination alone was not protective (except during an observed outbreak of PPR).