Multivariable BOLASSO regression of risk factors for campylobacteriosis.

<p>1. Multivariable BOLASSO logistic regression (odds ratio reported). 20-fold cross validation used for model selection.</p><p>2. Missing data was accounted for by multiple imputations using regularized regression (LASSO). 10 datasets were imputed and 100 bootstraps were then generated from each imputed dataset.3. PAF (Population attributable fraction) was calculated using odds ratios as approximations for risk ratios.</p><p>3. Significance indicators: 0</p><p>*** 0.001</p><p>** 0.01</p><p>+ 0.1.</p><p>Multivariable BOLASSO regression of risk factors for campylobacteriosis.