Skip to main content

Table 3 Description of the posterior classification from the BMI latent class model according to the baseline characteristics for MDR-TB patients

From: Different profiles of body mass index variation among patients with multidrug-resistant tuberculosis: a retrospective cohort study

 Slow BMI increaseRapid BMI increase 
Characteristics(n = 22)*(n = 121)*p-value
Age at baseline37.6 (13.8)32.6 (10.7)0.1002
Sex (female)15 (68.2)33 (27.3)0.0004
BMI (kg/m2)15.8 (2.5)17.9 (2.5)0.9989
Initial sputum smear (positive)19 (86.4)108 (89.3)0.2471
Initial sputum culture (positive)15 (79.0)104 (89.7)0.1156
HIV status (positive)13 (59.1)22 (18.3)0.0001
History of TB treatment (new case)9 (40.9)18 (14.9)0.0061
Treatment adherence (no)14 (63.6)4 (3.3)<.0001
Depression (yes)4 (18.2)1 (0.83)0.0002
Chest pain (yes)14 (63.6)55 (45.5)0.0550
Cough (yes)20 (90.9)120 (99.2)0.0586
Dyspnea (yes)15 (68.2)34 (28.1)0.0004
Nausea (yes)5 (22.7)7 (5.79)0.0175
Hemoglobin count9.8 (2.4)10.8 (2.1)0.3888
Platelet count377.7 (181.2)384.2 (133.5)0.0474
Lymphocyte count1.7 (1.6)1.9 (1.4)0.3289
Neutrophil count5.1 (3.0)4.9 (2.6)0.3083
Creatinine count79.5 (17.7)76.5 (20.2)0.4928
SGOT32.7 (9.5)29.2 (4.1)<.0001
SGPT34.2 (6.3)32.2 (4.7)0.0657
  1. Data are shown as the mean (SD) or number (%); SD standard deviation; bold p values were < 0.05, which means the corresponding factor was significantly associated with the BMI group latent class. Group membership was explained by treatment outcome and lung cavities on X-ray. *Lung cavities on X-ray at baseline was missing for 22 patients. Distributions of the baseline factors across these classes were compared a posteriori using a chi-squared test for the categorical variables and Student’s t-test for the continuous variables