The effect of geographical distance on TB patient delays in a mountainous province of China

Abstract

Setting: One hundred and twenty-nine counties in Yunnan, a mountainous province in China.

Objective: To document the relationship between patient delays and distance to local county tuberculosis (TB) centres.

Design: A computerised medical record-based study of a cohort of 10356 new smear-positive TB cases in 2005.

Results: The median total delay was 71 days (interquartile range [IQR] 38-128), with a median long patient delay of 60 days (IQR 28-111) and a relatively short median health care system delay of 4 days (IQR 2-13). Older age (>40 years), being an agriculturer and poor economic status were significantly associated with longer patient delays. Risk of delay increased with increasing geographical distance, with a greater effect on relatively shorter patient delays. Using the first quartile of distance as the reference group, hazard ratios for subsequent quartiles were 0.61 (0.57-0.65), 0.30 (0.28-0.33) and 0.15 (0.14-0.17) for short patient delays (<or=60 days), and 1.04 (0.94-1.17), 0.69 (0.63-0.77) and 0.43 (0.39-0.47) for long patient delays (>60 days).

Conclusion: Patients living in remote areas need support to overcome the barrier posed by geographical distance, which has a greater effect in the initial phases of the disease.