Data-quality issues and alternative variable-screening methods in a questionnaire-based study on subclinical Salmonella enterica infection in Danish pig herds
H. Stege et al., Data-quality issues and alternative variable-screening methods in a questionnaire-based study on subclinical Salmonella enterica infection in Danish pig herds, PREV VET M, 48(1), 2001, pp. 35-54
Our aim was to determine risk factors for subclinical Salmonella enterica i
nfection in Danish finishing-pig herds. In this paper, the evaluation, comb
ining and initial reduction of variables is presented, along with assessmen
t of the hypothesis in the preliminary statistical testing.
The first group of herds was selected at random with no former knowledge of
S. enterica infection. Both the herd prevalence and the within-herd preval
ence among these herds turned out to be low; hence, some additional herds w
ere selected from The Danish Salmonella Control program, based on their hig
h seroprevalence. This resulted in a hybrid case-"control" design of the st
udy and therefore, five different methods of categorising the data were use
d to ensure that variables were not wrongfully excluded as a result of usin
g an improper design.
Our questionnaire focused on management, infection-limiting precautions and
feed and feeding procedures. To establish the prevalence of S. enterica in
fection within herds at the time of the visit, 50 blood samples from each h
erd were collected and serologically examined. The reliability of each vari
able from the questionnaire was assessed and it was decided which variables
should be selected, disregarded, combined with other variables and/or reco
ded. In the simple statistical testing (2 x 2 tables. cut-off: P = 0.25) he
rds were defined as subclinically S. enterica infected if the within-herd p
roportion of individual pigs with OD%>10 was more than 20%.
The results included questionnaires from 96 randomly selected and 39 high-s
eroprevalence herds and 6814 blood samples. The initial 95 variables origin
ally included in the questionnaires were reduced to 21 by critical check, c
ombination, recoding and preliminary screening. We failed to demonstrate "h
erd size" as a risk factor for subclinical S. enterica infection in pig her
ds. (C) 2001 Elsevier Science B.V. All rights reserved.