Does union membership increase sickness absence from work and, if so, by how much? And which specific channels does this effect operate through? Using UK Labour Force Survey data for 2006-2008 we find that trade union membership is associated with a substantial increase in the probability of reporting sick and in the amount of average absence taken. This result can be largely attributed to the protection that unions offer to unionized employees. Supportive evidence is also found for a reduction in 'presenteeism' (attending work when sick) among union members. The results are robust to different modelling and estimation approaches.