Skip to main content
. Author manuscript; available in PMC: 2020 Aug 1.
Published in final edited form as: AIDS Care. 2019 May 9;31(8):965–972. doi: 10.1080/09540121.2019.1612020

Table 3:

Multivariate models predicting self-reported ART adherence with interactions by age

Multivariate Model, main effects only b (se) Multivariate Model with healthcare trust by age interaction b (se) Multivariate Model with ratings of doctor by age interaction b (se) Multivariate Model with problem drinking by age interaction b (se) Multivariate Model with HIV conspiracy beliefs by age interaction b (se)
Socio-Demographics:
 Age 50+ 4.50 (3.01) −21.90 (12.30)+ −34.20 (12.45)** −0.63 (3.68) 16.34 (9.04)+
 Low Income −8.19 (3.00)** −7.81 (2.97)** −7.84 (2.93)** −8.29 (2.96)** −8.28 (2.99)**
Mental Health:
 Depression −3.16 (3.81) −3.17 (3.78) −3.11 (3.73) −3.40 (3.77) −3.24 (3.81)
 Stimulant Use −8.18 (3.36)* −8.49 (3.33)* −8.42 (3.28)* −7.48 (3.33)* −7.56 (3.38)*
 Problem Alcohol Use −8.39 (3.11)** −8.96 (3.09)** −9.43 (3.06)** −13.64 (3.79)*** −7.88 (3.13)*
 Internalized HIV Stigma −3.78 (1.34)** −3.86 (1.32)** −3.48 (1.31)** −3.96 (1.32)** −3.91 (1.34)**
Trust/Mistrust:
 Perceived Art Efficacy 9.97 (2.91)*** 9.65 (2.89)*** 9.98 (2.85)*** 9.68 (2.88)*** 10.06 (2.91)***
 Trust in Healthcare 1.59 (2.17) −0.59 (2.37) 1.92 (2.13) 1.35 (2.15) 1.31 (2.18)
 Ratings of Doctor −0.02 (0.94) −0.31 (0.94) −1.94 (1.10)+ −0.20 (0.93) −0.15 (0.94)
 HIV Conspiracy beliefs 1.80 (1.72) 2.55 (1.74) 2.88 (1.72) 1.18 (1.72) 3.79 (2.24)+
 Social Support 0.50 (1.37) 0.77 (1.36) 0.51 (1.34) 0.62 (1.36) 0.46 (1.37)
Interactions:
 Trust in Healthcare by Age 7.73 (3.49)*
 Ratings of Doctor by Age 4.49 (1.40)**
 Problem Drinking by Age 15.20 (6.40)*
 HIV Conspiracy beliefs by Age −4.65 (3.35)

Note: Data were missing for some characteristics, ranging from n=0 to n=7 (3%). In such cases missing data were imputed with the overall mean. All models controlled for data source (Rise vs. Mednet), which was not significant for any model.

+

p < .10,

*

p <. 05,

**

p < .01,

***

p < .001.