Skip to main content
. 2013 Jun 9;18(2):323–334. doi: 10.1007/s10461-013-0494-0

Table 5.

Association of self-change factors with unsuppressed viral load among refugees and local host community on HAART for ≥25 weeks in Kuala Lumpur, Malaysia (N = 222)

Factor Prevalence ≥40 copies/mL, n/N (%) p value, crude odds ratio (95 % CI) p value, adjusted odds ratio (95 % CI)a
Adherence self-efficacy (self-rated ability to take medications as prescribed over previous month)b p = 0.37 p = 0.95
 Excellent 16/99 (16) 1 1
 Good/very good 14/105 (13) 1.30 (0.74, 2.26) 1.02 (0.56, 1.86)
 Very poor/poor/fair 6/18 (33)
Serostatus disclosure to partner p = 0.67 p = 0.77
 No 4/22 (18) 1 1
 Yes 17/120 (14) 0.74 (0.22, 2.46) 1.11 (0.29, 4.23)
 No partner 15/80 (19) 1.04 (0.31, 3.52) 1.45 (0.38, 5.53)
Serostatus disclosure to family/friends p = 0.23 p = 0.49
 No 10/81 (12) 1 1
 Yes 26/141 (18) 1.61 (0.73, 3.53) 1.37 (0.56, 3.34)
Alcohol use, past month p = 0.29 p = 0.69
 Never 24/164 (15) 1 1
 One or more times 12/58 (21) 1.52 (0.71, 3.28) 0.83 (0.33, 2.06)
Use of illegal/harmful substances, past six months p = 0.23 p = 0.83
 No 32/208 (15) 1 1
 Yes 4/14 (29) 2.20 (0.65, 7.45) 1.18 (0.27, 5.31)
Use of traditional medicines, past six months p = 0.46 p = 0.75
 No 29/188 (15) 1 1
 Yes 7/34 (21) 1.48 (0.57, 3.57) 1.31 (0.47, 3.70)
No. of reported barriers to adherenceb p = 0.46 p = 0.89
 0 13/82 (16) 1 1
 1+ 8/67 (12) 1.13 (0.82, 1.56) 1.03 (0.71, 1.49)
 3+ 8/36 (22)
 5+ 7/37 (19)
Knowledge of HIV and AIDS (% correct of four questions) p = 0.15 p = 0.23
 0+ 1/18 (6) 1 1
 50+ 35/204 (17) 3.52 (0.45, 27.33) 3.21 (0.37, 28.05)

Note: 32 clients with incomplete data were excluded (five missing viral loads; 13 missing pharmacy claim records). Clients with missing data were not significantly different (p > 0.05) from those retained for analyses on age, sex, refugee status, and time on HAART

aAdjusted for age group, sex, refugee status, travel in past year, time to clinic, time on HAART, time from HIV diagnosis to HAART start, and previous regimen switch

bFactor modelled as a linear effect (common odds ratios presented)