Tuberculosis national registries and data on diagnosis delay - Is there room for improvement?

Excessive delay in the diagnosis of Tuberculosis may have a negative impact on the epidemiological control and elimination of this disease. An accurate determination and analysis of delay times may help identify where and how to improve Tuberculosis diagnosis according to local needs. The Portuguese Tuberculosis Surveillance System - SVIG-TB - is the main source of data regarding diagnosis delay. However, to our knowledge, there has been no recent evaluation of its data. This study's primary aim was to conduct a thorough quantitative and qualitative evaluation of data obtained from the SVIG-TB registry concerning the delay in Tuberculosis diagnosis in Matosinhos, a Portuguese municipality.

All patients living in the Matosinhos municipality diagnosed with Tuberculosis between January 1^st 2012 and December 31^st 2019 were identified and individual SVIG-TB records retrieved. Patient-related, Healthcare-related and Total delay in Tuberculosis diagnosis were determined based on data obtained from this source and compared to data recovered from patient record review. Missing data, registering errors and differences in diagnosis delay between these two sources were analysed and compared. Subsequently, diagnosis delay data from a recent Portuguese nationwide SVIG-TB paper, covering years 2010 to 2014, was compared to local SVIG-TB and patient record review data.

This study identified a significantly greater percentage of cases with missing data on delay in SVIG-TB records when compared to Patient Record Review (57.2% versus 1.11%). The median patient-related, healthcare-related and Total delay in Tuberculosis diagnosis were of 17, 30 and 68 days, respectively, in the SVIG-TB based data. A significant underestimation of healthcare-related and total Tuberculosis diagnosis delay was identified in SVIG-TB data when compared to Patient Record Review. There was no significant difference between Patient-related delays determined from these two sources. Compared to the national study results, missing values were more common in this data set (57.2 vs 44.6%), mainly due to the absence of patient-related delay data. Median Total and Healthcare-related delays were significantly greater in Matosinhos Municipality, regardless of the data source (SVIG-TB or Patient Record Review). The patient-related delay was, conversely, shorter.

SVIG-TB has been crucial in guiding National Public Health policies on the path towards Tuberculosis elimination in Portugal. However, there is still room for improvement. These results provide a basis for further reflection on the shortcomings and potential of SVIG-TB in guiding the national Tuberculosis program.