Table 1 summaries the socio-demographic characteristics of the OST participants. Among the 231 clients, 113 had never used opioids via injection (non-IDU) and 118 injected opioids before enrollment into OST. There was a significant difference for age, with mean age of non-IDU clients being 32.15 years and 35.3 years for the IDU (p=0.001). There were few women in either group; 9.7% for non-IDU and 5.9% for IDU. Non-IDU were significantly better educated with about one in three (30.1%) educated up to university level and above but only 16.9% among the IDU (p=0.028). Significantly more IDU were separated or divorced 16.1% than non-IDU of 3.6% (p=0.002). About one in four (43.4%) non-IDU had attempted to quit substance use which was significantly less than the 74.6% IDU (p<0.001). The number of non-IDU with co-morbid medical conditions was significantly less (32.4%) than for IDU (67.6%) (p<0.001). More IDU belonged to upper middle socioeconomic class (61.9%) as compared to non IDU (45.8%) based on the modified Kuppuswami Scale for socio-economic status (p=0.014) .
Insert Table 1 about here
Table 2 outlines the difference in OST related characteristics between non-IDU and IDU clients. Methadone was the most used current OST modality in comparison to buprenorphine amid both non-IDU (80.5%) and IDU (69.5%). The duration of OST use was six or more months for most IDU clients (78.8%) and non-IDU clients (54%) (p=<0.001). Significantly less non-IDU (8.8%) were arrested or taken into police custody after OST enrollment as compared to IDU (26.3%) (p=0.001). The mean duration of OST enrollment was higher in the IDU group (40.6 months) compared to the non-IDU (13.1 months) (p<0.001).
Insert Table 2 about here
Table 3 illustrates the differences in QoL and psychological distress between non-IDU and IDUs. Statistically significant differences were observed in all QoL domains and the overall QoL with lower mean scores for all domains in IDU compared to non-IDU; physical QoL (p<0.001), psychological QoL(p=0.001), social QoL(p=0.004), environmental QoL (p=0.014), overall QoL (p<0.001). The highest mean difference was observed in the environmental domain with 28.4 for non-IDU group and 27 for IDU. The mean for overall QoL in non-IDU was 94.4 and 86.7 for IDU. Likewise, the mean of Kessler-6 score was significantly lower for non-IDU clients; 4.5 versus 7 for IDU (p=0.002), and similarly for SPD among 15.9% for non-IDU and 32.2% for IDU (p=0.004).
Insert Table 3 about here
Table 4 shows the unadjusted and adjusted odds ratios for factors associated with severe psychological distress (SPD) using the logistic regression model. The model was adjusted for IDU status, sociodemographic characteristics, history of psychiatric illness, co-morbidities, substance abuse history, and being in custody.
In the adjusted model, IDU status did not show significant association with the SPD. Those having history of psychiatric illnesses were around 7 times more likely to have SPD (AOR=7.19, CI:3, 17.22, p<0.001). Likewise, those with history of comorbidity also were around 1.6 times more like to have SPD than those who did not (AOR=1.58, CI: 1.11, 5.20, p=0.025).
Insert Table 4 about here
Table 5 presents the linear regression models for different domains of QoL adjusted for IDU status, sociodemographic characteristics, history of psychiatric illness, co-morbidities, substance abuse history, and being in custody. IDU were likely to have lower physical QoL [standardized Beta (β)= -0.21; 95% CI: -3.27, -0.75, p-value<0.01] compared to non-IDU. Age was also positively associated with better social QoL (β= 0.16; 95% CI: 0.01, 0.12, p-value<0.05).
A history of self-reported mental illness was associated with lower QoL on all four domains of QoL [physical domain (β= -0.30; 95% CI: -5.56, -0.2.48, p-value<0.001); psychological domain (β= -0.25; 95% CI: -4.25, -1.41, p-value<0.001); social domain (β= -0.28; 95% CI: -3.03, -1.16, p-value<0.001); and environmental Domain (β= -0.22, 95% CI: -4.22, -1.18, p-value<0.001)].
Similarly, previous quit attempt was also associated with lower QoL across all four QoL domains [physical domain (β=-0.14; 95%, CI:-2.53,-0.13, p-value<0.05); psychological domain (β=-0.19; 95% CI:-2.78,-0.57, p-value<0.01); social relationships (β=-0.13; 95% CI:-1.48,-0.04, p-value<0.05); and environmental domain (β=-0.14 95% CI:-2.48,-0.11, p-value<0.05)].
Insert Table 5 about here
Likewise, those with upper middle or above socio economic status were more likely to have good quality of life compared to ones belonging to the lower middle class socio economic status with higher QoL score across all domains of QoL [physical domain (β=0.14; 95% CI:0.09,2.65, p-value<0.05); psychological domain (β=0.23; 95% CI:0.76,3.11, p-value<0.001); social relationships (β=0.29; CI:0.88,2.43, p-value<0.001); and environmental domain (β=0.31; 95% CI:1.55,4.07, p-value<0.001)].