Analysis of the Influencing Factors of Immunological Nonresponders in Wuhan, China

Table 5

Multivariate linear regression analyses between Influencers and CD4⁺ cell count increase, stratified analyses by BMI.

Variable	BMI < 18.5	BMI 18.5–25.0	BMI ≥ 25.0
Gender	−2.2 (−83.5, 79.0)	−6.0 (−56.3, 44.2)	−38.9 (−211.5, 133.6)
Age at follow-up	0.4 (−1.2, 2.0)	−0.4 (−1.4, 0.6)	−1.5 (−4.6, 1.6)
Duration from HIV diagnosis to HAART start (months)^b	−24.6 (−58.5, 9.2)	−13.5 (−31.1, 4.1)	−50.7 (−138.3, 36.9)
Duration of HAART (months)^b	−5.0 (−43.3, 33.3)	−1.6 (−18.7, 15.5)	18.5 (−41.1, 78.0)
Transmission route	1.6 (−29.5, 32.6)	13.1 (−8.7, 34.9)	−32.4 (−133.7, 68.8)
WHO clinical stage	20.0 (−8.3, 48.4)	1.3 (−15.0, 17.6)	52.8 (−8.9, 114.6)
Co-trimoxazole^a	−93.7 (−157.8, 29.5)	−82.0 (−114.9, −49.1)	−595.7 (−756.2, −435.2)
HAART options^a	18.1 (−12.5, 48.6)	6.9 (−9.7, 23.5)	28.7 (−72.1, 129.4)
HAART replacement	−54.2 (−101.2, −7.2)	−5.2 (−36.0, 25.6)	−65.2 (−202.0, 71.6)
Cr^b	1.5 (0.1, 3.0)	0.7 (−0.2, 1.6)	−1.5 (−4.3, 1.3)

^aInfluencers in baseline, ^bInfluencers in follow-up, . The multivariate linear regression model included gender, age at follow-up, duration from HIV diagnosis to HAART start, duration of HAART, transmission route, WHO clinical stage, co-trimoxazole, HAART options, HAART replacement, cr.