Skip to main content
. 2021 May 7;16(5):e0251303. doi: 10.1371/journal.pone.0251303

Table 3. Characteristics of the study population with a history of Tuberculosis (TB) and stratified by the onset of the TB episode in relation to diabetes ascertainment.

ALL N = 1802 (98%) a TB episode before diabetes ascertainment N = 794 (44%) TB episode after diabetes ascertainment N = 1008 (56%) p-value
Sex (Female) 1023 (56.8%) 442 (55.7%) 581 (57.7%) 0.415
Age at diabetes ascertainment (Years) 49.0 [41.0;56.0] 49.0 [41.2;57.0] 48.0 [41.0;56.0] 0.13
Baseline HbA1c (%) 9.2 [7.1;11.8] 8.2 [6.8;11.0] 10.1 [7.6;12.4] <0.001
Baseline Hba1c < 7% 327 (22.3%) 189 (28.9%) 138 (17.0%) <0.001
Last HbA1c (%) 11.0 [7.6;14.1] 10.8 [7.2;14.4] 11.1 [8.0;13.9] 0.130
Last HbA1c < 7% 328 (18.2%) 173 (21.8%) 155 (15.4%) <0.001
HIV Positive 791 (43.9%) 410 (51.6%) 381 (37.8%) <0.001
Diabetes duration since ascertainment (Years) 5.1 [1.8;7.1] 3.3 [0.5;6.1] 6.1 [3.5;7.9] <0.001
Patient outcome (Deceased) 178 (9.9%) 53 (6.7%) 125 (12.4%) <0.001
Ever started TB treatment 1802 (100.0%) 794 (100.0%) 1008 (100.0%) .
Ever started diabetes treatment 1579 (87.6%) 640 (80.6%) 939 (93.2%) <0.001
Linkage to diabetes treatment b 1350 (85.5%) 536 (83.8%) 814 (86.7%) 0.120
Linkage to HbA1c testing c 1209 (67.1%) 532 (67.0%) 677 (67.2%) 0.983

a. 35 (2%) individuals who had ever had TB did not have enough data to classify when they had a TB episode relative to diabetes ascertainment

b. Proportions of patients having a record of diabetes drugs being prescribed within one year of their last recorded diabetes-related hospital encounter (calculated from those who had ever started diabetes treatment)

c. Proportion of patients having a recorded HbA1c laboratory test result within one year of the last recorded diabetes related hospital encounter