Determinants of patient and health care services delays for tuberculosis diagnosis in Italy: a cross-sectional observational study

We present a cross-sectional multicentre observational study enrolling all consecutive patients diagnosed with TB in seven referral centres in Italy. Information on demographic and clinical characteristics, health-seeking trajectories and patients’ knowledge and awareness of TB were collected. Diagnostic delay was assessed as patient-related (time between symptoms onset and presentation to care) and healthcare-related (time between presentation to care and TB diagnosis). Factors associated with patient-related and healthcare-related delays in the highest tertile were explored using uni- and multivariate logistic regression analyses.

We enrolled 137 patients, between June 2011 and May 2012. The median diagnostic delay was 66 days (Interquartile Range [IQR] 31–146). Patient-related and healthcare-related delay were 14.5 days (IQR 0–54) and 31 days (IQR: 7.25–85), respectively. Using multivariable analysis, patients living in Italy for < 5 years were more likely to have longer patient-related delay (> 3 weeks) than those living in Italy for > 5 years (Odds Ratio [OR] 3.47; 95% Confidence Interval [CI] 1.09–11.01). The most common self-reported reasons to delay presentation to care were the mild nature of symptoms (82%) and a good self-perceived health (76%). About a quarter (26%) of patients had wrong beliefs and little knowledge of TB, although this was not associated with longer diagnostic delay. Regarding healthcare-related delay, multivariate analysis showed that extra-pulmonary TB (OR 4.3; 95% CI 1.4–13.8) and first contact with general practitioner (OR 5.1; 95% CI 1.8–14.5) were both independently associated with higher risk of healthcare-related delay > 10 weeks.

In this study, TB was diagnosed with a remarkable delay, mainly attributable to the healthcare services. Delay was higher in patients with extra-pulmonary disease and in those first assessed by general practitioners. We suggest the need to improve knowledge and raise awareness about TB not only in the general population but also among medical providers. Furthermore, specific programs to improve access to care should be designed for recent immigrants, at significantly high risk of patient-related delay.

The study protocol was registered under the US National Institute of Health ClinicalTrials.gov register, reference number: NCT01390987. Study start date: June 2011.