Overall, 50,588 people participated in NHANES surveys from 2007 to 2016, of whom 1,878 participants who failed to complete medical exams and 23,799 participants who were younger than 18 or over 69 years of age were excluded. There were 24,911 participants eligible to answer the question “ever using cocaine, heroin, or methamphetamine” of whom 3,399 (13.6%) participants did not respond to this question. Participants with missing values were more likely to be female, younger in age, and belonging to other races. There were 1,283 participants who reported ever using methamphetamine, and among them 18.4% had HBV vaccine-induced immunity. After further excluding those with HBV vaccine-induced immunity and with no blood-borne infections (n = 208), 1,075 participants were eligible (Fig. 1) with 157 (14.6%) diagnosed with blood-borne infections and 918 (85.4%) susceptible to blood-borne infections. The number of participants infected by HBV, HCV and HIV were 95 (61%), 93 (59%), and 11 (7%), respectively. Additionally, among these infected by any blood-borne viruses, 6 were infected by both HIV and HBV, 36 were infected by both HBV and HCV, and no one was infected by both HCV and HIV. Table 1 summarizes the frequencies and weighted estimates for factors associated with blood-borne infections. Based on the weighted estimates, the 1,075 participants represented approximately 11,319,270 methamphetamine users in the U.S. population with the overall prevalence of blood-borne infections at 13.4 per 100,000. Specifically, the prevalence of infection by HBV, HCV and HIV were 7.1, 7.8 and 1.1 per 100,000, respectively.
In the study sample, a third of the participants were female, about two thirds (64%) were non-Hispanic white, 25% of the participants were 50 to 59 years old and they accounted for 41% of blood-borne virus infections. Approximately, a quarter (23%) of the participants were living below the poverty threshold and they accounted for 35% of blood-borne infections; another 28% were living between 1 to 1.9 times poverty index and they accounted for 30% of blood-borne infections. A third of the participants did not have any health insurance, and nearly a quarter (23%) did not have routine healthcare access. About 30% of the participants with blood-borne virus infections had less than high school education. -About 2% were identified as men who had sex with men (MSM) and they accounted for 9% of blood-borne infections. While only 19% of methamphetamine users also reported ever using the other two drugs (heroin or cocaine), they accounted for over half (52%) of blood-borne infections. The majority of methamphetamine users did not inject any drugs (78%); however, almost two thirds (65%) of blood-borne infections were among the 22% of persons with IDU. Approximately 44% of participants first started using methamphetamine at age 18 to 25, and 22% between 10 and 17 years, and another 22% when they were older than 25 years.
Table 2 summarizes the estimated model effects (uOR and aOR with 95% CIs) of factors associated with the outcome variable. From bivariate analysis, being 50 to 59 years old (uOR 3.98; 95% CI 1.14–13.91), being 60 to 69 years old (4.05; 1.34–12.22), being non-Hispanic black (2.07; 1.15–3.71), living around poverty index 1 to 1.9 (2.07; 1.17–3.67), living below the poverty threshold (3.02; 1.66–5.49), having lower than high school education (2.23; 1.31–3.81), identified as MSM (17.45; 5.39–56.55), using methamphetamine with other two substances (5.93; 2.50–14.05), IDU (6.36; 3.84–10.54), having used methamphetamine more than 50 times (2.69; 1.39–5.24), and started using methamphetamine at age over 26 (2.05; 1.16–3.61) were statistically significantly associated with blood-borne infections among people who reported using methamphetamine.
The three-step multivariable logistic regression models further adjusted for all the covariates sequentially. The effect size of all aORs with 95% CIs are illustrated in Fig. 2. In model 1, only being older than 50 was significantly associated with blood-borne infections. After adding socio-economic status and sexual activities into model 2, there were significant associations of blood-borne infections with age 50–59 years old (aOR 8.94; 95% CI 2.09–38.26), living around poverty index 1 to 1.9 (2.89; 1.45–5.77), living below the poverty threshold (3.78; 1.61–8.90), having lower than high school education (3.24; 1.60–6.57), and identified as MSM (25.25; 6.78–94.00). In the final model 3, after adding drug use behaviors, associations of the same risk factors with blood-borne infections persisted: age 50–59 years old (6.32; 1.35–29.69), living around poverty index 1 to 1.9 (2.80; 1.33–5.88), living below the poverty threshold (2.46; 1.14–5.28), having lower than high school education (3.57; 1.74–7.33), and identified as MSM (54.24; 13.80–213.24). In addition, in this model, using methamphetamine with other two substances (5.86; 1.50–22.87), IDU (3.77; 1.93–7.36), and started using methamphetamine at age over 25 (2.18; 1.05–4.54) were also significantly associated with blood-borne infections.