INTRODUCTION

Tobacco smoking is very common among HIV infected people, estimated at approximately 50 to 70 percent¹. In Nepal, 47% of HIV infected people were current tobacco smokers². Mortality and morbidity were reduced and life expectancy increased among HIV infected people after the introduction of antiretroviral therapy (ART)³. In addition, non-AIDS related cardiovascular disease and malignancies were reported to be the main causes of death amongst people living with HIV (PLHIV) compared to the general population^4-6. However, death caused by the side effects and the long-term toxicity of ART amongst PLHIV remain unclear⁷. Smoking tobacco among PLHIV alters immune and virological responses, leading to increased vulnerability to infection, tuberculosis and low adherence of ART^8,9. On the other hand, life expectancy was found lower among tobacco smoking PLHIV than in non-smoking PLHIV¹⁰. Tobacco products were associated with higher mortality among PLHIV compared with just being infected with HIV¹¹. Unemployment, alcohol use, illicit drugs, low ART adherence, low social support, low education level, decreased CD4 cells count and discriminations were found significantly associated with tobacco smoking amongst PLHIV¹².

Tobacco smoking increased among PLHIV with alcohol consumption¹³. Drinking of alcohol was higher among PLHIV than the general population¹⁴. Alcohol consumption accounted for more deaths among PLHIV than by only HIV and other diseases¹⁵. Consumption of alcohol produced a toxicity that reacted with ART, resulting in elevated risk of liver diseases, co-infections, development of other diseases, as well as absorption and metabolism of drug and immune dysfunction¹⁶. Non-adherence to ART, decline of CD4 cells count, increased viral load, rapid development of diseases and delayed on care of HIV were the factors associated with drinking alcohol^17,18. In addition, risky sexual behavior was also linked to drinking alcohol, which increased the risk of HIV transmission¹⁹.

The HIV epidemic is high in developing countries. Tobacco and alcohol related health risks are more common among PLHIV. Very few studies are available in Nepal related to the problems of PLHIV. Thus, we aimed to determine the proportion of people who smoke tobacco and drink alcohol, as well as assess the factors associated with tobacco smoking and drinking of alcohol, amongst PLHIV in Nepal.

METHODS

RESULTS

The mean (SD) age of the participants was 36.1 (7.8) years. The majority (53.0%, 95% CI: 44.4-61.6) of the participants were female. Most of the participants followed the Hindu religion (70.5%, 95% CI: 61.9-78.1) and were of non-indigenous ethnicity (56.1%, 95% CI: 47.5-64.6) (Table 1). One-third (36.4%, 95% CI: 28.2-45.2) of the participants had not attained formal education. Most of the participants (84.8%, 95% CI: 78.6-91.0) had children and among them 79.5% (95% CI: 70.8-86.5) had ≤ 2 children. Monthly per capita income was low, with 64.4% (95% CI: 55.6-72.5) of the participants earning≤ 5000 NPR (1USD=100NPR) (table not shown).

Of the total HIV infected people (PLHIV), onefourth (26.5%, 95% CI:18.9-34.1) were current smokers and one-fifth (22.7%, 95% CI: 15.5-30.0) were current alcohol drinkers (Table 1). The majority of participants (56.8%, 95% CI: 47.9-65.4) were diagnosed with HIV infection at an age ≤ 34 years; while 54.5% (95% CI: 45.6-63.2) of participants were found taking ART for less than a year after being diagnosed HIV positive (table not shown). Similarly, the majority of participants 54.5% (95% CI: 45.6-63.2) reported that they were infected from sex workers; 65.0% (95% CI: 56.3-73.6) of the participants’ spouses were HIV infected; 75.8% (95% CI: 68.3-83.2) of the participants were sexually active; 16.0% (95% CI: 7.1-18.9) had sexual contact with someone other than the spouse; and one-fourth of the participants (26.0%, 95% CI: 17.2-34.7) had sexual intercourse without using a condom (Table 1).

The morbidity of tuberculosis among HIV infected participants was 13.6% (95% CI: 7.7-19.6). Around half of the participants (52.3%, 95% CI: 43.4-61.0) were diagnosed as less than or equal to second clinical stage and the average (SD) CD4+T-cell count was 300.6 (SD=158.3) (table not shown). Two-fifths of the respondents (41.7%, 95% CI: 33.1-50.6) had ever forgotten to take ART medication (Table 1).

Table 2 shows significant differences between tobacco smoking and alcohol drinking with demographic, behavioral and health characteristics. Tobacco smoking was significantly higher among HIV infected males. Tobacco smoking was significantly different by occupation. Tobacco smoking was significantly different between alcohol drinkers and non-drinkers. The highest proportion of tobacco smoking was found among the PLHIV who were infected after sexual contact with sex workers, whereas a lower proportion was found amongst the respondents who had a HIV positive spouse; but a slightly higher proportion was found among those who had sexual contact with their infected spouse.

Alcohol use was significantly different by religion, tobacco smoking, and duration of ART medication. The highest prevalence of alcohol drinking was found among the respondents who were infected after sexual contact with sex workers; slightly higher prevalence was among those who had sexual contact with their spouse; and lower prevalence was among those who had sexual contact using condoms. Statistically significant difference was found with those variables. Alcohol drinking was higher among HIV infected individuals who ever forgot to take ART medication (Table 2).

Table 3 shows logistic regression results for tobacco smoking and alcohol drinking. Male participants were more likely to smoke tobacco (OR=12.19, 95% CI: 4.32-34.38, p<0.001) than females, and participants with formal employment were more likely (OR=3.95, 95% CI: 1.36-11.47, p=0.012) to smoke tobacco than those who were unemployed. The participants who were infected after sexual contact with sex workers were more likely (OR=15.2, 95% CI: 4.35-53.08, p<0.001) to smoke tobacco than those who were infected by other than sexual contact. The participants who lived with a HIV infected spouse were less likely (OR=0.10, 95% CI: 0.04-0.27, p<0.001) to smoke tobacco than those who lived with a HIV negative spouse.

Participants who followed the Hindu religion were less likely (OR=0.31, 95% CI: 0.13-0.72, p=0.007) to drink alcohol compared with those who followed another religion than Hinduism. Participants who were taking ART medication for more than or equal to a year were more likely (OR=2.57, 95% CI: 1.11-5.96, p=0.028) to drink alcohol than those who were taking medication for less than a year. Participants who were infected after sexual contact with sex workers were more likely (OR=4.50, 95% CI: 1.70-11.93, p=0.003) to drink alcohol than those who were infected by other than sexual contract. HIV infected individuals who used a condom during sexual contact were less likely (OR=0.10, 95% CI: 0.03-0.27, p<0.001) to drink alcohol. HIV infected individuals who ever forgot to take ART medication were three times more likely (OR=3.17, 95% CI: 1.36-7.38, p=0.008) to drink alcohol than those who never forgot to take ART medication (Table 3).

DISCUSSION

The prevalence of tobacco smoking and alcohol drinking among HIV infected people was found to be higher than the general population. Our study revealed that the proportion of tobacco smoking (26.5%) and alcohol drinking (22.7%) among HIV infected people is high.

Male participants revealed the significant association with current tobacco smoking, followed by formal employment, and mode of HIV infection after having sexual contact with sex workers. Our study revealed a higher proportion of tobacco smoking among male PLHIV compared to the national rate²⁰. One of the possible reasons of the low prevalence among female respondents could be under reporting in the data by females, which might not replicate the exact prevalence. The degree of under reporting of smoking status by females is unknown in low income countries with strong social and cultural apprehensions against tobacco smoking by females. Our findings revealed current smoking was associated with formal employment, contrary to previous findings that revealed that current smoking was associated with unemployment²¹. Most of those with formal employment can purchase tobacco more easily, which could be a possible reason, while HIV status might result in a weak economic situation resulting in a lack of money to purchase tobacco, as would also be among the unemployed. However, the result was a bit contradictory as the tobacco smoking prevalence was higher among the low-income group, though not statistically significant. Similarly, previous studies indicated that lower socioeconomic groups were more likely to smoke tobacco²¹. Current tobacco smoking had significantly high prevalence among those who were HIV infected from sex workers. Psychological and depressive states might be the possible reason for being more likely to smoke among PLHIV²². Tobacco smoking was low among PLHIV when both members of the couple were HIV infected. If the couple had a similar problem, they might have adjusted themselves to circumstances to avoid depression.

Alcohol drinking was more likely to be associated with current tobacco smoking among PLHIV. This finding indicated that either smoking or alcohol drinking encouraged an individual to trigger co-addiction, among HIV infected individuals. Previous literature revealed that tobacco smoking rate among HIV infected people was higher when someone smoked along with alcohol consumption^13,23,24.

The longer the duration of ART medication after HIV diagnosis the more likely the drinking of alcohol, which might be due to more side effects at the beginning of ART medication. The feeling of uncertainty to be cured after ART medication and mental distress might be other reasons to indulge in alcohol drinking²⁵. Our findings revealed that prevalence of alcohol drinking was significantly higher among the respondents who were HIV infected from sex workers. Psychological problems could be the main reason for drinking alcohol and persist after feeling of regret²⁵. Previous studies have revealed a relationship between alcohol drinking and risky sexual behavior^19,26. Another study revealed that sexual intercourse without a condom was associated with a higher likelihood for alcohol consumption among HIV infected people²⁷. Our results are in agreement with these findings. Risky sexual behavior increased vulnerability to HIV transmission as well as to other infections. ART medication interruptions were related to the possibility of treatment failure²⁸. Our study findings revealed non-adherence of ART medication was associated with alcohol drinking. Stopping ART medication might be because of the fear of interactive toxicity with alcohol that one felt after drinking¹⁶. Previous literature highlighted that non-adherence to ART, decline of CD4 cells count, and delayed on care of HIV were associated with alcohol consumption^17,18,29.

This study has several strengths and limitations. We used probability sampling of HIV infected individuals; and assessed the self-reported information related to substance use and ART medication adherence. Due to the small sample size, we could not include different controlling variables in the regression model. Our findings were based on cross-sectional data and small sample size that limited us from drawing causal inferences. Obtaining information about sensitive behaviors like drug use through the face-to-face interview process could lead to underreporting as a result of social desirability biases. We did not collect initiation time of substance use and the quantity of the substance used. We did not use standardized tools to assess tobacco smoking, alcohol drinking and ART adherence. We did not measure the frequency of alcohol drinking and tobacco smoking, therefore we were unable to perform further analysis. The sample was extracted from a central level ART clinic and the participants were selected from those clients who visited there from across the country to get the services. Therefore, the findings of this study could not be generalized to other HIV infected populations in other settings. However, this study could simplify a significant proportion of substance use among HIV infected people in Nepal to some extent.

CONCLUSIONS

The proportion of tobacco smoking and alcohol drinking is high among HIV infected people compared to the general population in Nepal. Low ART medication adherence is associated with alcohol drinking. There is an urgent need to develop tobacco cessation and alcohol avoidance cost-effective interventions for HIV infected people. Integrated counseling, tobacco cessation and motivational programs are needed in ART clinics and community settings, which might be helpful to improve the ART medication adherence and help HIV infected people to live longer.