Long-term outcomes of a cell phone-delivered intervention for smokers living with HIV/AIDS

Abstract

Background: People living with human immunodeficiency virus (HIV)/AIDS (PLWHA) have a substantially higher prevalence of cigarette smoking compared to the general population. In addition, PLWHA are particularly susceptible to the adverse health effects of smoking. Our primary objective was to design and test the efficacy over 12 months of a smoking cessation intervention targeting PLWHA.

Methods: Participants were enrolled from an urban HIV clinic with a multiethnic and economically disadvantaged patient population. Participants received smoking cessation treatment either through usual care (UC) or counseling delivered by a cell phone intervention (CPI). The 7-day point prevalence abstinence was evaluated at 3, 6, and 12 months using logistic regression and generalized linear mixed models.

Results: We randomized 474 HIV-positive smokers to either the UC or CPI group. When evaluating the overall treatment effect (7-day abstinence outcomes from 3-, 6-, and 12-month follow-ups), participants in the CPI group were 2.41 times (P = .049) more likely to demonstrate abstinence compared to the UC group. The treatment effect was strongest at the 3-month follow-up (odds ratio = 4.3, P < .001), but diminished at 6 and 12 months (P > .05).

Conclusions: Cell phone-delivered smoking cessation treatment has a positive impact on abstinence rates compared to a usual care approach. Future research should focus on strategies for sustaining the treatment effect in the long term.

Keywords: HIV/AIDS; cell phone intervention; smoking cessation.

Figure 1.

Consolidated Standards of Reporting Trials (CONSORT) diagram, showing screening, study enrollment, and retention through 12-month follow-up.

Figure 2.

Smoking abstinence by treatment group, showing the percentage of participants in each treatment condition (usual care vs cell phone intervention) who reported 7-day smoking abstinence at the 3-, 6-, and 12-month follow-up periods. Bars represent the standard error of the mean. A and B, Percentage abstinent using the intent-to-treat and complete-case approach, respectively.