TABLE 3.

Association between RD genotype and extrathoracic TB assessed by logistic regression models by use of the GEE method without and with adjustment for HIV serostatus, gender, race/ethnicity, and plcD genotype

SubgroupRD genotypeCrude OR (95% CI)aAdjusted OR (95% CI)b
1Without deletion1.001.00
2RD105, RD181, and RD1422.46 (1.42, 4.26)3.05 (1.58, 5.90)
3RD105, RD181, and RD15012.33 (4.72, 32.23)11.09 (4.27, 28.80)
4RD105 and RD1811.49 (0.34, 6.54)2.21 (0.58, 8.44)
5RD105c
6Partial RD105c
7RD2391.03 (0.22, 4.76)1.18 (0.22, 6.36)
  • a Results from model 1, including only the RD genotype.

  • b Results from model 6, including the RD genotype as well as HIV serostatus, gender, race/ethnicity, and plcD genotype.

  • c The ORs of RD105 deletion alone and a partial RD105 deletion were not calculable because all patients infected by an isolate with an RD105 deletion alone or by an isolate with a partial RD105 deletion had thoracic TB.