The numbers behind that headline are technically striking. For 7-day point-prevalence abstinence, abstinence during the seven days immediately preceding a follow-up visit, the pooled relative risk was 2.07 (95% CI: 1.60–2.68), with essentially no statistical heterogeneity across the eleven contributing trials (I² = 0.0%). Translated into absolute terms, the pooled risk difference of 0.086 corresponds to an increase of roughly 8.6 percentage points in quit rates, meaning that about twelve people living with HIV would need to receive a telehealth-integrated intervention for one additional person to quit smoking in the short term. The 30-day point-prevalence abstinence outcome, reported by three trials, told a consistent story: a relative risk of 2.04 (95% CI: 1.31–3.19), an absolute gain of about 5.2 percentage points, and a number needed to treat of approximately twenty.

Why does this matter clinically? Around 36.1% of people living with HIV currently smoke, a rate two to three times higher than in the general population. Although antiretroviral therapy has dramatically reduced deaths from immunodeficiency, all-cause mortality among people living with HIV remains substantially elevated compared with age- and sex-matched individuals, and tobacco and alcohol exposure are major contributors to that excess. Long-term smoking and drinking can offset the survival benefits of antiretroviral therapy and raise the risk of cardiovascular disease, liver damage, and multiple cancers. When smoking and drinking co-occur, as they often do in this population, their synergistic effects may exacerbate immunosuppression and amplify mortality risk further.

Telehealth, in this context, refers to remote behavioral support delivered through information and communication technology: proactive telephone counseling, automated text messaging, web-based cessation platforms, smartphone applications, and interactive voice response systems. The review’s authors argue these channels address barriers that have long undermined in-person interventions for people living with HIV, including the risk of identity exposure and the stigma that weakens patients’ willingness to engage, the considerable time costs of frequent clinic trips, and the uneven regional distribution of intervention resources. A pragmatic randomized controlled trial across thirteen HIV outpatient clinics in Vietnam had already demonstrated that telehealth-delivered tobacco cessation support was clinically feasible, and supportive text messaging has been shown to enhance abstinence maintenance and reduce total alcohol consumption.

The evidence base was assembled rigorously. The team searched PubMed, Embase, the Cochrane Library, and Web of Science from database inception to April 1, 2026, screening 2,161 records down to fifteen eligible randomized controlled trials, all prospectively registered with PROSPERO (CRD420251242171) and reported under PRISMA 2020 guidelines. Thirteen trials came from the United States, one from China, and one from Cambodia. Eleven trials targeted HIV-positive smokers, most of whom smoked at least five cigarettes per day, while four trials focused on heavy drinkers or individuals meeting diagnostic criteria for alcohol use disorder. Smoking outcomes were primarily verified objectively through exhaled carbon monoxide thresholds and salivary cotinine, and intention-to-treat data were preferentially extracted.

The interventions themselves were diverse. Telephone-based programs, the most successful modality in exploratory subgroup analysis with a relative risk of 2.99 (95% CI: 1.67–5.38), involved eight to eleven proactive follow-up calls grounded in cognitive behavioral therapy and motivational interviewing, sometimes supplemented by a counseling hotline. Combined telephone-plus-text-message programs reached a relative risk of 2.57 (95% CI: 1.35–4.91), and web-based interventions achieved 1.62 (95% CI: 1.12–2.35). Text-message-only programs (RR = 1.95, 95% CI: 0.89–4.28) and app-based interventions (RR = 2.46, 95% CI: 0.76–7.99) favored the intervention but did not reach statistical significance. Most smoking cessation packages were paired with nicotine replacement therapy or varenicline, so the pooled estimates capture the average incremental effect of entire intervention packages, not the isolated effect of the technology itself.

One exploratory finding deserves particular attention: the intervention effect estimate was significantly larger at shorter follow-up. In studies assessing abstinence at three months or less, the relative risk reached 2.82 (95% CI: 1.92–4.14), compared with 1.53 (95% CI: 1.07–2.18) in studies with longer follow-up, a statistically significant subgroup difference (p = 0.02). The authors frame this through the Transtheoretical Model of behavior change: in the early action phase, external reminders such as calls and texts may actively support behavior modification, whereas long-term maintenance may demand sustained self-efficacy and the establishment of healthy behavioral patterns to replace smoking. Reduced self-regulatory capacity and continued exposure to stress or smoking-related cues may raise relapse risk, underscoring the potential importance of sustained support. Importantly, because this was a study-level comparison, it does not prove that intervention effectiveness necessarily declines over time.

The alcohol results were more sobering. Across four trials involving 352 participants, telehealth-integrated interventions, mostly built on the HealthCall platform in interactive voice response or smartphone app form, delivering daily behavior monitoring with monthly visual feedback and motivational interviewing, were associated with a reduction in mean drinks per drinking day (MD = −0.33, 95% CI: −0.58 to −0.08). However, no clear between-group difference emerged in the number of drinking days in the past 30 days (MD = −0.02, 95% CI: −0.35 to 0.30). Each pooled alcohol outcome rested on only three studies, all alcohol outcomes were self-reported through instruments such as the Timeline Followback rather than biochemically verified, and in sensitivity analyses excluding the study rated at high risk of bias, the effect estimate for mean drinks per drinking day remained in the same direction but its confidence interval crossed the null. GRADE assessments rated the certainty of evidence as low for all three alcohol-related outcomes and for 30-day abstinence, while the 7-day abstinence evidence achieved the highest rating.

Methodological quality was moderate. Using the Cochrane RoB 2.0 tool, three trials were classified as low risk overall, ten raised some concerns, and two were rated high risk. Missing outcome data performed best, with all fifteen trials rated low risk, while outcome measurement and selective reporting were the principal weaknesses: seven trials lacked publicly available trial registration or clearly reported analysis plans, leaving the possibility of selective outcome reporting unresolved. Publication bias for the 7-day abstinence outcome appeared absent, with a generally symmetrical funnel plot and Begg’s test yielding p = 0.186, and leave-one-out and random-effects sensitivity analyses left every pooled estimate materially unchanged, with the random-effects relative risk for 7-day abstinence at 2.00 (95% CI: 1.54–2.59).

The review’s limitations shape how far its conclusions travel. Thirteen of fifteen trials were conducted in the United States, leaving regions with the highest HIV burden, including Africa, much of Asia, and South America, essentially unrepresented. Exploratory meta-regression found no significant associations between the 7-day abstinence effect and study-level mean age, proportion of male participants, or sample size, though statistical power was limited. The authors also caution that the absence of statistically significant differences across telehealth modalities (p = 0.45) should not be read as evidence of equivalent efficacy, and that the app subgroup contained only one trial in which just 9.5% of participants used the intervention app daily and consistently. Telephone counseling’s direct human interaction may help sustain motivation, yet apps and text messages retain irreplaceable scalability advantages in resource-limited settings. The bottom line for clinicians and researchers alike: telehealth-integrated packages may meaningfully improve short-term smoking cessation among people living with HIV and may modestly reduce drinking intensity, but adequately powered randomized controlled trials with longer follow-up, objective corroboration of alcohol use where feasible, and enrollment in high-burden settings are needed to confirm durability and generalizability.

Subject of Research: Effects of telehealth-integrated interventions on smoking cessation and alcohol reduction among people living with HIV

Article Title: Effects of telehealth-integrated interventions for smoking cessation and alcohol reduction among people living with HIV: a systematic review and meta-analysis

Article References: Yang, P., Yi, J., Asai, G., & Zhao, F. (2026). Effects of telehealth-integrated interventions for smoking cessation and alcohol reduction among people living with HIV: a systematic review and meta-analysis. Frontiers in Public Health, 14, Article 1962757. https://doi.org/10.3389/fpubh.2026.1962757

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1962757

Keywords: telehealth, smoking cessation, HIV, alcohol reduction, meta-analysis, systematic review, randomized controlled trials, point-prevalence abstinence, mHealth, public health, Effects, telehealth-integrated