BackgroundWhile user fees in healthcare systems have been associated with quality improvement, a substantial increase may have a detrimental effect. This paper reports on the effects of increasing user fees on utilisation of TB diagnostic services in Tanzania.MethodsWe retrospectively analysed data on TB diagnostic services utilisation between July 2013 and June 2015 in Mnazi Mmoja Zanzibar (MMZ), Musoma and Sumbawanga hospitals. In July 2014, user fees in Musoma were increased substantially from 2 to 5 US dollar; Sumbawanga increased the fees stepwise, from 1 to 2 US dollar in July 2014, and from2 to 3 US dollar in January 2015 MMZ did not raise the fees. We compared TB services utilisation before and after introduction of user fees.ResultsOut of 7483 presumptive TB patients registered in all sites, 50.2% were males. Over half (3969) were registered before the user fee was increased. Among 3969, 1579 (39.8%) were from Musoma, 922 (23.2%) from Sumbawanga and 1468 (37.0%) from MMZ. Of the 3514 patients registered after the introduction of user fees, 983 (28%), 952 (27.1%) and 1579 (44.9%) patients were from Musoma, Sumbawanga and MMZ, respectively. The number of presumptive TB patients seeking TB diagnostic services at Musoma decreased significantly by 38% from 1579 to 983 after the increase of user fees (p=0.001). More females (817; 51.8% vs 458, 35.9%) attended Musoma before user fees were increased as compared to males whose attendance did not differ much (761; 48.2% vs 525; 53.4%); (p=0.01). There was no significant decrease of patients at Sumbawanga and MMZ.ConclusionThere was a significant decrease in the number of presumptive TB patients who accessed new TB diagnostic services in Musoma after a substantial increase of user fees, the effect was stronger among women. Although user fees are beneficial, they should be increased stepwise so as not to affect service utilisation.