Geographic pattern of tuberculosis diagnosis among patients in two regions of Cameroon

• Main Authors: Ching-I Lu, 盧靖宜
• Other Authors: 林先和
• Format: Others
• Language: en_US
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/jv4quq

碩士 === 國立臺灣大學 === 流行病學與預防醫學研究所 === 107 === Background There is a long journey from getting tuberculosis (TB) to recovery. Repetitive visits to TB centers for diagnosis and treatment is common in high-burden settings. During this period, the geographic barriers can be one important determinant of whether the patient could eventually complete the treatment. Few studies used geographic information systems to assess accessibility to TB service. The present study aims to evaluate the geographic barriers of TB diagnosis and treatment and their determinants in Cameroon. Methods We estimated the travel distance for all notified TB patients (n=1870) from 39 out of 40 diagnostic and treatment units in the northwest and southwest regions of Cameroon between 2015 and 2016. Travel distance of notified TB patients was estimated by network analysis in ArcGIS using the road information in the study area. We calculated two types of travel distances, one based on the distance between the patient’s address to the nearest TB center (as a proxy measurement of geographic barrier), the other based on the distance between the patient’s address to the TB center that the patient actually visited. A “cross-center” visit is defined if a patient did not visit the nearest TB center and the difference between the two types of distances was greater than 5 kilometers. We used logistic regression analyses to evaluate the factors of cross-center visit. Results Based on 39 tuberculosis diagnostic and treatment unit locations, the median travel distance to the nearest TB center was 5.23 kilometers (IQR=13.14), but the median travel distance to the TB center actually visited was two times greater (11.82 kilometers, IQR=27.21). 609 out of 1946 (31.3%) TB patients had cross-center visit. In the univariable logistic regression analysis, region of the DTU (Southwest vs. Northwest, OR=0.60, 95%CI: 0.50-0.74), type of the DTU (faith-based vs public, OR=3.72, 95%CI: 3.04-4.57), marital status (married vs single, OR=1.38, 95%CI: 1.03-1.84) are the significant factors associated with cross-center visit. Conclusions To improve accessibility to TB control facilities in areas, the establishment of TB centers is important, but not enough. The prevailing phenomenon of cross-center visit identified in the present analysis suggested that there were hidden factors (probably associated with quality of care) that added to the barriers to receiving TB care. Efforts are needed urgently to understand the reasons of cross-center visit in order to improve the efficiency of TB diagnostic and treatment network.