BACKGROUND: The Dutch contact investigation guidelines stipulate that Public Health Services should examine contacts around all pulmonary tuberculosis (TB) patients to prevent disease and further transmission. Our objective was to assess to what extent these guidelines were applied and whether patient characteristics were associated with having contacts investigated. METHODS: We extracted the records of all reported pulmonary TB patients from the nationwide surveillance register covering 2006-07. Patient characteristics associated with having contacts investigated were assessed by multivariable logistic regression analysis. RESULTS: Out of the 1236 pulmonary TB patients reported, 909 (74%) patients were eligible for analysis, since 133 (11%) patients had incomplete records and 194 (16%) patients were registered by Public Health Services who did not report contact investigation results. For 710 (78%) out of the 909 patients contacts were investigated. Compared with Dutch patients, contacts were significantly less often investigated around immigrant patients (84 vs. 75%, OR: 0.60; 95% CI: 0.40-0.92). Contacts were significantly more often investigated for smear positive patients (OR: 3.52; 95% CI: 2.23-5.55) and culture positive patients (OR: 2.71; 95% CI: 1.76-4.16), compared with smear negative and culture negative patients, respectively. CONCLUSION: Initiating contact investigations appear to be prioritized based on the infectiousness, but also on the ethnicity of pulmonary TB patients. By not investigating the contacts of 25% of the immigrant patients, there is a risk of missing a significant number of infected and diseased contacts, since the incidence in this group is markedly higher than in the Dutch population.