Abstract: In the United States more than half of everyone who ever smoked has quit. Most people addicted to nicotine require several quit attempts to stop, but in some people with HIV a failed quit attempt predicts future success. People with HIV appear to have a harder time quitting than people without HIV. CDC analysis of nationally representative samples figured a quit ratio of 52% in the general population versus 32% in HIV-positive people. Studies in HIV populations indicate that those most likely to quit smoking include older people, pregnant women, and people with a high motivation to quit, a previous quit attempt, or recent pulmonary disease. US health authorities recommend that clinicians adopt the 5As approach to smoking cessation: ask, advise, assess, assist, and arrange. But only a little more than half of US clinicians assist smokers in picking a smoke-ending strategy, and only 10% arrange follow-up within the first week after a quit date. Smoking-cessation medications recommended and tested in people with HIV are varenicline, bupropion, and nicotine replacement in various forms. Success with these strategies generally ranges from 10% to 20% in HIV-positive people, with higher success rates in some subgroups. Successful nondrug strategies to support drug therapy in people with HIV include an Internet-based interactive program, cell-phone reminder calls, one-time 1-hour one-on-one counseling, and formal clinician education.
More people have quit smoking in the United States than still smoke.1 And more than half of everyone who ever smoked has stopped.1 Those eye-opening statistics from the Surgeon General may be the best way to challenge hardcore smokers -- and reluctant clinicians -- who say quitting is just too tough. The Surgeon General's findings make it clear that many can -- and do -- break their dependence on nicotine.
Who Quits Smoking and Who Starts Again?
Research reviewed in this article suggests that several standard strategies -- and perhaps a few novel approaches -- have helped HIV-positive people quit smoking. A steadfast few can stub out their last butt, resolve never to light up again, and free themselves for life. But most face a sterner challenge. And that challenge may prove greater for people with HIV. Analysis of a nationally representative sample of US residents with HIV ranked 42.4% as current smokers, 20.3% as former smokers, and 37.3% as never smokers.2 In the general-population National Health Interview survey, only 20.6% were current smokers, while 21.9% smoked in the past and 57.5% never smoked. Centers for Disease Control and Prevention (CDC) researchers calculated that the generalpopulation group had a quit ratio* of 51.7%, compared with 32.4% in the HIV group.
And quitters can become relapsers. A 1994-2011 study of 2961 HIV-positive and 981 HIV-negative women in the Women's Interagency HIV Study (WIHS) found that smoking prevalence in this largely black and Hispanic cohort waned from a high of 57% in 1996 to a low of 39% in 2011.3 Among 1622 women who smoked at their first study visit, 316 (19.5%) quit for a hefty median of 16.5 years. But among 273 sustained quitters with follow-up data, 145 (53%) resumed smoking after an average 7.2 years without cigarettes.
Quitting and relapsing rates proved similar in a Swiss HIV Cohort Study (SHCS) analysis that included 4833 HIV-positive smokers, about one quarter of them women. Through 2012, 1261 smokers (26%) quit at least temporarily.4 Among 1167 quitters with follow-up data, 557 (48%) resumed smoking. CDC researchers turned up several clues to which HIV-positive smokers are more or less likely to quit by analyzing quit ratios in a nationally representative sample of people in care for HIV infection in 20092 (Table 1). Quitting rates track with age, as people 50 or older -- more concerned about mortality -- had the highest quit ratio (42.6%) while 18- to 29-yearolds had the lowest (16.3%). Women with HIV had a substantially lower quit ratio than HIV-positive men (27.0% versus 34.2%), and blacks had a lower quit ratio (26.2%) than Hispanics (35.4%) or whites (38.2%). Quitting proved significantly more likely in HIV-positive people with more than a high school education (quit ratio 39.6%) than in those who only completed high school (26.8%) or didn't finish high school (25.1%). HIV-positive people at or above the poverty level were significantly more likely to quit than people below the poverty levels (quit ratio 39.9% versus 24.3%). For every category listed, people with HIV had quit ratios significantly below those of smokers in the general population comparison group.
Table 1. HIV-Positive People Likely to Quit Smoking or Relapse After Quitting
More Likely to Quit
Less Likely to Quit
More Likely to Relapse
Not on cART3
Previous quit attempt4
CD4 count <200 cells/mm3 (vs >500)
Smoked more daily before quitting4
High motivation to quit4
Women vs men2
Poor motivation to quit4
Recent pulmonary disease5
Blacks vs Hispanics or whites2
Not living in one's own home3
Crack, cocaine, or heroin use3
Having health insurance3
Poor self-reported health3
Recent hospital admission4
Nationally representative US sample,2 Women's Interagency HIV Study,3 Swiss HIV Cohort Study,4 Veterans Aging Cohort Study.5
cART, combination antiretroviral therapy.
Analysis of 2961 HIV-positive women and 981 demographically similar HIV-negative women in WIHS identified several factors independently linked to longer time to quit smoking: less than high school education, only high school education (versus more), having health insurance, smoking for more than 10 years, 1 to 6 alcoholic drinks weekly or more (versus none), fair to poor self-reported health (versus good to excellent), and hypertension (Table 1).3 Limiting the analysis to women with HIV, the WIHS team pinpointed four additional risk factors for longer time to quitting smoking: household income at or below $12,000 yearly, use of combination antiretroviral therapy (cART), CD4 count below 200 cells/mm3 (versus 500 or higher), and use of crack, cocaine, or heroin. Pregnant women were more likely to quit smoking sooner.
To single out predictors of quitting and relapse, SHCS investigators built two models.4 The smoking cessation model classified smokers into highly, poorly, and typically motivated people based on common-sense predictors of high motivation (like a previous quit attempt or a recent cardiovascular diagnosis) (Figure 1). The relapse model incorporated nicotine dependence level (maximum cigarettes smoked daily before quitting), the three motivational groups, age, and gender.
The Swiss team found that smokers who stopped were more likely to be highly motivated and to have tried stopping before.4 Smokers were less likely to stop if they were poorly motivated, had a recent hospital admission, had psychiatric comorbidities, or had a history of alcohol or drug dependence.
A Veterans Aging Cohort Study (VACS) team probed for predictors of quitting in 1027 HIV-positive and 794 HIV-negative veterans in care between 2005 and 2007.5 More than half of both groups, 56%, had tried to quit before, and two thirds of both groups were contemplating another attempt to quit. Among HIV-positive smokers, those with recent pulmonary disease proved almost 5 times more likely to have tried quitting recently (adjusted odds ratio [aOR] 4.93, 95% confidence interval [CI] 1.41 to 17.17). Unhealthy alcohol use cut odds that HIV-positive or negative veterans were contemplating quitting (aOR 0.66, 95% CI 0.49 to 0.90 for HIV positive).
Certainly, clinicians should not refrain from encouraging patients to stop smoking because they have one or more predictors of lower quitting likelihood. For example, the SHCS study found people with psychiatric comorbidity less likely to stop smoking.4 But the WIHS study of mostly black and Hispanic US women found no evidence that depression made women less likely to quit or more likely to relapse.3 People who inject drugs may seem poor candidates for smoke-ending interventions, but some are readier than others. A study of 267 HIV-positive current or former injection drug users who smoked found that older age and having a supporter who used medication to quit smoking doubled the odds of using smoke-ending medication.6
Among HIV-positive and negative WIHS women who stopped smoking for more than 12 months, marijuana use and not living in one's own home independently predicted a shorter time to resumed smoking (Table 1).3 Limiting the analysis to HIV-positive women identified two more predictors of a shorter time to relapse -- enrollment in 2001-2002 (versus 1994-1995) and use of crack, cocaine, or heroin. Older age and cART use predicted a longer time to relapse. In the SHCS study, relapsers smoked more in the three follow-up visits before quitting and had poor motivation to quit (Figure 1).4
Clinicians should tell smokers who quit and relapse that many people try to quit several times before succeeding, so they shouldn't give up if the first few tries fail. Previous quit attempts predicted later success in the Swiss HIV Cohort Study.4
The WIHS team speculates that HIV-positive women with a CD4 count below 200 cells/mm3 may be less likely to stop smoking because their more advanced HIV disease presents more immediate health challenges and robs them of motivation.3 That may also explain why women taking cART (who probably had more advanced disease) were less likely to quit in that analysis. But this WIHS study also linked taking cART to a longer time to relapse among women who quit at least 12 months. That could mean women doing well on cART have more motivation to improve their overall health, the authors propose. The surprising finding that women with health insurance took a longer time to quit, the WIHS team suggests, may mean women without insurance worry more about behaviors that threaten their health. But the insurance finding seems to be at odds with the link between lower income and lower probability of quitting.
The SHCS investigators found that their complex statistical models did no better than simple models (Figure 1) in picking HIV patients more likely to quit. "As a rough rule of thumb for clinicians," they suggest, "patients in our highly motivated group, especially those with a history of attempting to stop, and those known to have stopped recently are the best candidates for an intervention."4
First Steps in Getting HIV Patients to Quit
US health authorities urge clinicians to adopt the 5As approach to smoking cessation (Figure 2).7 Research indicates that primary care providers do a good job on the first two As -- ask every patient whether they smoke and advise smokers to stop. But dwindling proportions of clinicians implement the more timeconsuming third, fourth, and fifth As. That falloff became clear in a study of 3336 people enrolled in the National Lung Screening Trial, which found lower lung cancer mortality in participants randomized to screening by low-dose computed tomography than in those randomized to chest x-ray.8 Participants reported that three quarters of their providers asked them about smoking and advised them to stop (Figure 3).9 But fewer than two thirds of providers assessed whether patients were willing to quit, only 56% assisted them in picking a smoke-ending strategy, and a mere 10% arranged follow-up after the patient began to quit.
Veterans Administration (VA) clinicians may not represent all HIV providers in the United States, but a VA study indicates that clinicians caring for veterans often do not know which patients smoke, and this gap in care is worse with HIV-positive veterans.10 The study compared reports of 801 HIV-positive and 602 HIV-negative veterans with reports of 72 HIV providers and 71 non-HIV providers. While 82% of non-HIV providers correctly identified current smokers, only 65% of HIV providers knew which of their patients smoked.
Models adjusted for age, gender, race, and other factors determined that having HIV doubled the odds that the provider failed to identify current smoking (aOR 2.28, 95% CI 1.48 to 3.53).10 And compared with general practitioners, infectious diseases specialists proved almost 3 times more likely to miss current smokers (aOR 2.71, 95% CI 1.79 to 4.11). Providers did not identify current smokers more often when patients reported "bothersome" shortness of breath or cough, or when patients had smoking-related diseases like chronic obstructive pulmonary disease (COPD), coronary artery disease, or bacterial pneumonia. While 62% of non-HIV providers felt confident about influencing smoking cessation, only 39% of HIV providers did (P = 0.049).
A reasonable way to start encouraging HIV-positive people to forsake smoking is to count off the key benefits of quitting. The CDC highlights several major documented benefits of quitting, and research involving people with HIV confirms these benefits and adds others:
Overall benefits of quitting smoking
Smoking-related mortality plunges up to 90% in smokers who quit when younger than 40.1
When smokers over 50 years old quit, their excess cardiovascular disease risk disappears within 5 years.11
Two to 5 years after quitting, risk of stroke may fall to the same level as someone who never smoked.12
Within 5 years of quitting, risk of several cancers -- mouth, throat, esophagus, and bladder -- drops by half.12
Ten years after quitting, lung cancer risk drops by half.12
HIV-specific benefits of quitting smoking
Overall mortality risk is lower in former smokers than current smokers with HIV.13
Risk of cardiovascular disease, including myocardial infarction, is lower in former smokers than current smokers13 and decreases with longer time since quitting.14
Non-AIDS cancer risk is lower in former smokers than current smokers with HIV.13
Giving up cigarettes cuts risk of bacterial pneumonia, a frequent and serious complication of HIV infection.13,15-17
In the interview in this issue, HIV smoking expert Jonathan Shuter underlines the importance of building a quit plan before trying to stop smoking. The National Cancer Institute's SmokeFree.gov offers a step-by-step interactive guide to help people design their personal quit plan. Creating a quit plan involves seven steps:
Set a quit date.
List your reasons for quitting.
Identify smoking triggers you should avoid.
Build a "quit kit" to help you fight tobacco cravings.
Get rid of smoking reminders.
Pick a smoke-ending medication with your provider and line up support.
Tell friends and family you plan to quit.
See "Ten Things Every HIV-Positive Smoker Should Know" and "Ten Things Every HIV Clinician Should Know About Smoking" in this issue for summaries of long-term smoking risks and quitting benefits in this issue. At the same time, clinicians should not overlook the short-term benefits of quitting that may spark the interest of patients who want to see immediate gains from their abstinence. In a useful review article, Barcelona HIV experts Marta Calvo-Sánchez and Esteban Martínez note that short-term gains include having more money, tasting food better, and having better skin and less dyspnea.18
Smoke-Ending Drug Therapy Options
In their comprehensive review of smoking and HIV, Barcelona clinicians Calvo-Sánchez and Martínez note "significant resistance among the general population and the HIV-positive community" against taking drugs to help quit smoking.18 Fewer than half of HIV clinicians responding to a nationwide survey reported frequently prescribing antismoking drugs.19 But relying on will power alone to quit, they observe, results in repeated failure to escape nicotine addiction and in quit rates below 10%.
Calvo-Sánchez and Martínez list three first-line medication options -- various forms of nicotinereplacement therapy, varenicline (Chantix), and bupropion (Zyban).18 They relegate nortriptyline and clonidine to second-line status because of their side effects. These HIV smoking experts bluntly advise that "before a smoking cessation attempt, one of the first-line drugs should be prescribed in combination with health advice."18 The reason for this exhortation is simple: These drugs work. The Cochrane Database of Systematic Reviews lists 53 trials of nicotine replacement gum in the general population, 41 trials of replacement patches, and 14 studies of replacement tablets, inhalers, or nebulizers.18 Cochrane counts 19 bupropion trials in the general population, 7 varenicline trials, 6 clonidine trials, and 4 nortriptyline trials. With each of these options, odds ratios for successful cessation versus comparison interventions range from 1.43-fold (nicotine gum) to 2.33-fold (varenicline), and odds of success are always significant.
By blocking nicotinic acetylcholine receptors, varenicline blunts the craving for nicotine. Results of two small pilot trials20,21 and a larger placebocontrolled trial22 found at least short-term success with varenicline in people with HIV. The first study involved22 people who took varenicline for an average 30 days.20 Abstinence could be confirmed by cotinine levels in 6 people (27%) after 3 months and in 5 (23%) after 6 months. One person stopped varenicline after 1 week because of vomiting, headache, and discomfort. In an open-label nonrandomized study of 36 people averaging 29 pack-years of smoking (for example, 2 packs a day for 14.5 years), 6 people (17%) stopped varenicline because of side effects.21 No one had grade 3 or 4 lab abnormalities or serious side effects. Fifteen of these 36 people (42%) had cotinine-confirmed continuous smoking abstinence in weeks 9 to 12 of treatment.
ANRS, the French national HIV trials group, conducted the placebo-controlled trial in 250 HIV-positive people who smoked at least 10 cigarettes daily and said they wanted to quit.22 The researchers excluded people with current psychoactive drug dependence, a suicide attempt, or ongoing depression. US prescribing information warns of "serious neuropsychiatric events," including depression and suicide, in some people using varenicline.23 HIVsmoking mavens Calvo-Sánchez and Martínez suggest that "depression or psychiatric disorders at baseline are not contraindications [to varenicline] but great caution is recommended when depressive symptoms appear."18
Participants in the placebo-controlled trial smoked a median of 20 cigarettes daily, and more than 80% tried to quit at least once before.22 Their age averaged about 45, and 83% were men. Enrollees started varenicline at a dose of 0.5 mg once daily on days 1 to 3, followed by 0.5 mg twice daily on days 4 to 7, then two 0.5-mg tablets twice daily through week 12. The researchers confirmed self-reported quitting by measuring exhaled carbon monoxide. All participants received counseling on quitting throughout the study.
A modified intention-to-treat analysis involving 102 people who started varenicline and 111 who started placebo calculated continuous abstinence from weeks 9 through 12 in 34.3% randomized to varenicline and 12.6% randomized to placebo, rates that yielded 3.6-fold higher odds of quitting with varenicline (95% CI 1.8 to 7.2, P = 0.0003).22 From week 9 through week 48, the quit rate dwindled to 17.6% with varenicline versus 7.2% with placebo, results yielding still nearly tripled odds of quitting with varenicline (OR 2.8, 95% CI 1.1 to 6.7, P = 0.024). Grade 3 or 4 adverse events developed in 25% in the varenicline arm and 13% in the placebo group. Grade 3 or 4 drug-related adverse events affected 10% taking varenicline, while grade 3 or 4 drug-related psychiatric events affected 6%.
A nonrandomized US study of 228 HIV-positive smokers found that varenicline outdid nicotinereplacement therapy in helping people stop smoking24 as handily as it outdid placebo in the French trial.22 Study participants had enrolled in the Lung HIV study of respiratory complications; they smoked at least five cigarettes daily and expressed interest in quitting in the next 30 days. Researchers encouraged them to try varenicline for 12 weeks, and those who could not or opted not to received nicotine replacement therapy. Everyone got 12 weeks of telephone counseling. The investigators confirmed 7-day abstinence by saliva cotinine levels or exhaled carbon monoxide.
Age averaged 43, and 85% of participants were men.24 While 118 took varenicline, 110 used a nicotine patch (with gum as needed). Rates of confirmed abstention at the end of treatment were 25.6% with varenicline and 11.8% with nicotine replacement, results that translated into 2.54-fold greater odds of abstention with varenicline (95% CI 1.43 to 4.49). About 14% taking varenicline switched to nicotine replacement because of adverse events, which included nausea, abnormal dreams, agitation, insomnia, and depression. One person taking varenicline reported suicidal ideation, a problem that resolved when varenicline stopped.
Bupropion may help people stop smoking via two mechanisms. First, the drug selectively inhibits reuptake of the addiction drivers dopamine and norepinephrine. Bupropion also blocks nicotine receptors. Prescribing information for bupropion warns that people taking the drug may feel hostility, agitation, or depression and may have suicidal thoughts.25 Protease inhibitors and efavirenz may lower bupropion concentrations.
Because bupropion is also prescribed as an antidepressant, Weill Cornell Medical College HIV experts call it "an attractive option" for some HIV-positive smokers.26 But bupropion remains poorly studied as a smoking stopper in people with HIV. In a Spanish case series of 21 HIV-positive smokers who took bupropion, 8 (38%) gave up cigarettes for more than 1 year.27 No one had clinically significant interactions with antiretrovirals.
Nicotine replacement remedies now come in five formats -- patch, gum, nasal spray, oral inhaler, and lozenge. Scores of studies in the general population confirm that tapering smokers off nicotine with nicotine replacement can help them quit smoking.18 In the larger studies of nicotine replacement in people with HIV, quit rates have been as high as 26%.
Researchers at six US clinics randomized 444 HIV-positive smokers to 8 weeks of nicotine patch replacement plus standard care (two brief counseling sessions) or to nicotine patch plus more intensive motivation (four counseling sessions plus a quitday call).28 Study participants smoked an average 18 cigarettes daily, 63% were men, 52% European American, 18% African American, and 16% Hispanic. Almost two thirds had at least a high school education, but 79% were unemployed.
Intention-to-treat analysis determined carbon monoxide-confirmed 7-day abstinence rates of 13% at 2 months after enrollment and 10% at 4 months and 6 months in the standard care group versus 12% and 9% in the enhanced-counseling group.28 Six-month abstinence rates were significantly higher in Hispanics, people with low nicotine dependence, and people with high motivation to quit. Statistical analysis adjusted for these factors determined that more patch use boosted abstinence odds by one third (OR 1.32, 95% CI 0.99 to 1.75).
Another US study involved 209 HIV-positive smokers in care at three clinics and randomized to 10 weeks of nicotine patch or gum plus (1) individual counseling, (2) Internet-based counseling, or (3) self-help.29 Study participants averaged 45 years in age, 82% were men, 53% white, 27% black, and 14% Hispanic. More than three quarters of participants (79%) had at least a high school education, and 35% had education beyond high school. Only 14% had a job, while 11% were retired. They smoked an average 18 cigarettes a day.
About one quarter of participants in each study arm achieved carbon monoxide-verified 7-day abstinence 12 weeks after treatment began.29 One year after treatment began, quit rates remained near that level -- 20% with self-help, 20% with individual counseling, and 26% with Internet-based counseling, and these differences between groups were not significant. Employment, greater desire to quit, and lower mood disturbance scores favored quitting.
Which Smoke-Stopping Drug to Use
In the general population nicotine replacement therapy plus other interventions yielded 1-year quit rates of 15% to 31%30 -- results not too different from those recorded in the 209-person HIV trial.29 In the Cochrane Database review by Calvo-Sánchez and Martínez, nicotine replacement therapy efficacy compared with control arms ranged from 1.43-fold higher with gum to 2.02-fold higher with nasal spray.18 Odds of success with bupropion were similar (1.94-fold), while odds of success with varenicline were slightly higher (2.33-fold). In a nonrandomized study in Madrid involving 12 smokers who took bupropion, 8 who used nicotine replacement, and 6 who took varenicline -- all with four to six psychological support sessions -- 12-month quit rates were similar with the three interventions: 25% (3 of 12) with bupropion, 25% (2 of 8) with nicotine replacement, and 17% (1 of 6) with varenicline.31
University of Alabama at Birmingham researchers devised and tested an algorithm to help HIV clinicians figure whether varenicline, bupropion, or nicotine replacement suits an individual patient best.32 Providers should find and copy this useful algorithm in the cited article.32 The algorithm rests on a few simple decision drivers:
Does patient want to quit smoking today?
Yes → go to next decision point.
No → nicotine replacement plus
Is patient willing to take an oral medication twice daily?
Yes → consider contraindications to varenicline, then bupropion (Table 2).
No → consider contraindications to nicotine replacement (Table 2).
Contraindications to one medication drive patients to another medication.
Has patient unsuccessfully tried that medication before?
Yes with varenicline → consider bupropion.
Yes with bupropion → consider nicotine replacement.
Yes with nicotine replacement → prescribe two nicotine replacement methods.
Has patient quit after 4 weeks on medication?
No with varenicline → varenicline plus nicotine replacement.
No with bupropion → bupropion plus nicotine replacement.
No with one nicotine replacement → two nicotine replacements.
Table 2. Pluses and Minuses of Three First-Line Smoke-Ending Medications
Contraindications and Cautions*
Quit Rate in General Population
Quit Rate in People With HIV
Suicide attempt, depression or psychiatric disorders, kidney impairment, pregnancy
* Adapted from Calvo-Sánchez and Martínez,18 Cropsey,32 and product information.
The algorithm study involved 100 adults in care for HIV who smoked at least 5 cigarettes daily and were not pregnant or nursing.32 They got randomized to algorithm-driven treatment or treatment as usual (smoking cessation assistance from their provider when they were ready to quit). All participants received one standard 20-minute smoke-ending session in which staff discussed behavioral methods to reduce smoking. The authors picked varenicline as the preferred oral agent because of an eight-trial general-population analysis demonstrating its superiority to bupropion or placebo.33
The 100 study participants averaged 46 years in age; 71% were men, 75% black, and 25% white; 70% had at least a high school education.32 Two thirds of participants had received treatment for substance abuse and 57% for mental health. Enrollees smoked an average 16 cigarettes daily and had smoked for an average 27 years. The algorithm group and the usualcare group did not differ in any of these measures. Of the 50 people randomized to the algorithm group, 38% used the nicotine patch, 36% varenicline, 12% nicotine patch plus lozenge, and 10% bupropion. Twenty-six people randomized to the algorithm cluster had not tried to quit in 4 weeks and were offered the nicotine lozenge to ease acute cravings.
More people randomized to the algorithm than to usual care took a smoke-ending medication (81% versus 23%, P < 0.001). Through 16 weeks of follow-up, people in the algorithm group lowered their smoking rate more than people in the control group (10 versus 6 fewer cigarettes daily, P = 0.021). Throughout follow-up, a higher proportion in the algorithm arm made at least a 24-hour quit attempt (50% versus 38%, P = 0.006). The researchers argue that this algorithm "provides clinicians with a brief and succinct pathway for treatment selection decisions and thereby lends itself to being utilized during even brief clinical encounters."32
Are E-Cigarettes an Option for Smokers With HIV?
Makers of e-cigarettes (electronic nicotine delivery systems, conveniently shortened to ENDS) promote their ability to ease nicotine addiction and thus put smokers on the road to quitting. A few studies (none of them in people with HIV) found that e-cigarettes can help people reduce nicotine intake,35 perhaps on a level equivalent to nicotine patches.36 But the potential value -- and risks -- of e-cigarettes remain largely unknown.
E-cigarettes have three parts: a cartridge containing a liquid solution of nicotine and other chemicals, a heating element that vaporizes the liquid, and a power source, usually a battery (Figure 4).37 Proponents of e-cigarettes say they are safer than cigarettes because don't produce tar and other chemicals that come from burning tobacco. But e-cigarette vapors do contain substances other than nicotine, including some carcinogens and toxic substances such as formaldehyde, acetaldehyde, and derivatives of benzene and benzodiazepine.18,37
The FDA does not regulate e-cigarettes but the agency has begun to study them. Quality-control of processes used to make e-cigarettes, the FDA found, "are substandard or non-existent."38 Cartridges labeled as containing no nicotine did contain nicotine, and three cartridges with the same label produced markedly different nicotine levels. The FDA sent warning letters to three e-cigarette makers accusing them of unsubstantiated claims and poor manufacturing processes.
The National Institute on Drug Abuse cautions that "very little data exists on the safety of e-cigarettes, and consumers have no way of knowing whether there are any therapeutic benefits or how the health effects compare to conventional cigarettes."37 Calvo-Sánchez and Martínez say "reasons for skepticism about the potential benefits of ENDSs" include "paradoxical hindering of tobacco smoking cessation, inhalation of propylene glycol, unexpectedly high plasma nicotine levels, and inadequate information about the contents of ENDSs and the chemical environmental contamination they may produce."18
Nondrug Smoke-Ending Strategies for People With HIV
As HIV smoking maven Jonathan Shuter observes in the interview in this issue, writing a prescription for smokers and wishing them good luck are not enough. Medication-based smoking cessation attempts must be backed by counseling and some form of structured support. The most basic non-drug support may be provider follow-up within the first week of a scheduled quit attempt, as called for in the 5As approach to smoking cessation (Figure 2),7 or referral to a smoke-ending support line. A national quit line (1-800-QUIT-NOW) routes callers to state smoke-ending call lines (see Quit by Phone box below). Nondrug support can be considerably more extensive than that. Most of the drug intervention trials reviewed above included at least one nondrug element. This article reviews four nondrug strategies that complement pharmacotherapy and have been studied in people with HIV: an Internet-based approach smokers can start today, a cell-phone reminder strategy, a physician-training program, and a one-session counseling strategy.
Quit by Phone: 1-800-QUIT-NOW
A national quit line, 1-800-784-8669, refers callers to state phone lines that may provide an array of services:
Free support, advice, and counseling from experienced quitline coaches
A personalized quit plan
Practical information on how to quit, including ways to cope with nicotine withdrawal
The latest information about stop-smoking medications
Free or discounted medications (available for at least some callers in most states)
PositivelySmokeFreeMe (PSFM) is an 8-session 7-week interactive program that any smoker with Internet access can start today.39 In the pilot trial described here, 18% of HIV-positive smokers who completed all 8 sessions stopped smoking. Almost one third of women who viewed all 8 sessions quit. All participants were offered the nicotine patch and could use other cessation-aiding drugs.
The study took place at Montefiore Medical Center in the Bronx, which serves more than 2800 HIV-positive people in a poor to middle-class part of New York City. Between March 2012 and April 2013, the Montefiore team recruited HIV-positive smokers who said they wanted to quit and had access to an Internet-linked computer. They excluded pregnant women and people with low English or Spanish literacy. Participants got randomized to standard care (under 5 minutes of counseling and a self-help brochure) or to PSFM. Everyone got a prescription for a 3-month supply of nicotine patches, and everyone had full insurance for patch therapy. PSFM includes 8 separate sessions of 4 to 7 interactive Web pages that become available over 7 weeks. These pages aim to educate, motivate, and increase self-efficacy to quit smoking. The primary efficacy endpoint was carbon monoxide-verified 7-day abstinence 3 months after participants were supposed to complete the 8 sessions.
Among 138 people who entered the study, 134 completed follow-up, including 68 of 69 in the PSFM group. Age averaged 46, about 40% of participants were women, three quarters black, and almost half Hispanic. Although 90% had stable housing, 86% were unemployed. These people smoked an average 11 cigarettes daily. None of these factors differed significantly between study groups.
Of the 51 study participants who used any smokeending drug, 24 were in the PSFM group and 27 in the control group. Two thirds of people randomized to PSFM visited six or more online sessions, 41% visited all eight, and one third viewed all 41 Web pages. Almost everyone in the PSFM group, 94%, needed phone calls reminding them to view their next Web session.
In an intention-to-treat analysis, the 3-month 7-day quit rate measured 10.1% in the PSFM arm versus 4.3% in the control arm, a nonsignificant difference (Figure 5). But among 28 Internet users who logged into all 8 sessions, the quit rate came to 17.9%. Quit rates were higher for women overall than men (11.7% versus 2.7%, P = 0.08) and higher still for women who logged into all 8 sessions (30.8%) and women who viewed all 41 Web pages (40%) (Figure 5).
Although adherence in this trial required frequent phone-call reminders, adherence might be better in a population that spends more time online and enjoys interactive programs. Because PositivelySmokeFreeMe is free and immediately available, HIV providers may consider recommending it to motivated smokers. A new and enhanced version of PositivelySmokeFreeMe, including a moderated social network, will be online in the first half of 2016. (See the interview with Jonathan Shuter in this issue for more details about PositivelySmokeFreeMe.)
Cell-Phone Reminder Calls
Three months of counseling via cell phone improved 3-month quit rates compared with usual care in a randomized trial of HIV-positive smokers in Houston.40 But that advantage waned over the next 9 months and overall quit rates were low, perhaps because few study participants used nicotine replacement therapy, as recommended.
The study involved 474 adult smokers willing to set a quit date within 7 days. Researchers randomized 238 people to standard care (written smoke-ending materials and instructions on how to get nicotine patches at the study clinic) and 236 to the cell-phone intervention (standard care plus a prepaid cell phone, proactive smoking counseling calls for 3 months, and access to a support hotline).
The cell-phone group was significantly younger than the control group (43.9 versus 45.7 years), but the groups were balanced in proportion of women (about 30%), blacks (about 75%), whites (about 12%), Hispanics (about 10%) and in years of education (about 11), employment (about 20%), not working because of illness (about 63%), cigarettes smoked daily (about 19), illicit drug use in the past 30 days (40%), and major depressive symptoms (67%). Three quarters of participants in both groups completed 12 months of follow-up.
In an intention-to-treat analysis about 12% in the cell-phone group and 4% in the control group quit at 3 months.40 Among people who completed all 12 months, respective quit rates at 3 months were about 15% and 5%. Quit rates remained stable through 6 and 12 months in the control group while dropping in the cell-phone group to levels equivalent to control participants. Multivariate analysis determined that people randomized to get cell-phone reminders had quadrupled odds of quitting at 3 months (OR 4.3, 95% CI 1.9 to 9.8, P < 0.001). Through 12 months chances of quitting remained significantly higher in the cell-phone arm (OR 2.41, 95% CI 1.01 to 5.76, P = 0.049).
The researchers suggested high nicotine dependence coupled with low nicotine patch use could explain the low quit rates in both study arms. Trial investigators did not give participants nicotine patches but instead told them how to get patches through the clinic. But county requirements that people getting patches take smoke-ending classes may have discouraged patch use in this study group. The waning impact of cellphone counseling after 3 months, the authors suggest, could mean a longer intervention may be needed for smokers like these.
Clinician Training Program
Instead of making smokers the initial target of cessation efforts, Swiss HIV Cohort Study (SHCS) researchers in Zurich aimed at HIV providers in a half-day session on patient smoking counseling and drug therapy.41 This nonrandomized study comparing Zurich smokers with smokers at other SHCS sites found that those cared for in Zurich were about 25% more likely to stop smoking and 25% less likely to relapse.
All physicians at the University Hospital Zurich HIV outpatient clinic took a half-day smoking-cessation course between November 2007 and December 2009. The course included information on identifying smokers, rating nicotine dependence and motivation, counseling, and prescribing drugs to help smokers quit. During the study period these physicians completed a checklist documenting the smoking status of each patient and physician support offered (counseling, explaining medications available, arranging followup appointments, and if appropriate setting a quit date). The SHCS investigators defined quitting as a study visit in which a patient smoked followed by at least two consecutive visits without smoking. Relapse meant consecutive study visits in which a patient had quit followed by a visit with smoking.
The analysis involved 11,056 SHCS cohort members across the country, including 1689 seen at the Zurich clinic during the intervention period. The smoking cessation analysis considered 5805 smokers with at least three follow-up visits, while the relapse analysis involved 1953 smokers who quit. Zurich cohort members were older than SHCS members at other centers (median 44 versus 38), and Zurich had a lower proportion of women (25% versus 34%). Zurich and other cohort members smoked a median of 20 pack-years. Except for Zurich, none of the other SHCS centers offered structured smoke-ending programs for clinicians. About 80% of Zurich clinicians counseled smokers during the intervention period, a result indicating that such counseling can fit into routine care.
Overall smoking prevalence in SHCS members plunged from 60% in 2000 to 43% in 2010. The overall drop was greater in Zurich (-22.5%) than in other SHCS centers (-16.5%) or private practices (-14.5%). Logistic regression analysis determined that during the provider training period Zurich smokers were almost 25% more likely to quit smoking than SHCS smokers at other centers (OR 1.23, 95% CI 1.07 to 1.42, P = 0.004), and Zurich quitters were 25% less likely to relapse (OR 0.75, 95% CI 0.61 to 0.92, P = 0.007). The effect of the intervention was stronger than the effect of calendar time (OR 1.19 versus 1.04 per year).
The Zurich team proposed that "our approach of an institution-wide training programme for infectious diseases physicians to improve smoking cessation counselling can be well integrated into routine HIV care, was well accepted by patients and physicians, and can support patients' efforts to stop smoking."41 A 2008 survey of 363 HIV clinicians across the United States found that only 23% ever had formal tobacco treatment training.19 Only 29% of these US clinicians said they would be interested in attending a brief smoking cessation session, another 29% said they might be interested, and 40% claimed no interest. The Zurich HIV group now provides clinicians a yearly 1-hour smoking cessation class. Other SHCS centers around the country have not adopted that practice.42
One-Hour One-Time Smoker Counseling
A 1-hour individualized smoke-ending session with a physician, other health professional, or peer helped 16% of HIV-positive smokers in Newark quit through 6 months, a rate similar to results reported in other studies of more complex and time-consuming strategies.43
The Infectious Disease Practice in Newark, New Jersey43 cares for a largely minority, low-income HIV population similar to those in the Bronx39 and Houston40 studies. The Newark analysis involved HIV-positive adult smokers interested in quitting. Prospective participants were referred to the smoking program by their clinic provider, by clinic staff, or by themselves in response to fliers in waiting rooms and exam rooms. The hour-long counseling session with a physician, case manager, peer navigator, or mental health counseler (all trained in smoking cessation) focused on practical problem solving, withdrawal symptoms, and coping strategies. Clinicians prescribed smoke-ending therapy either at the counseling session or at a follow-up visit. Participants self-reported smoking cessation at 6 months at a follow-up visit or by phone.
Among 1545 adults cared for at the HIV clinic, 774 (50%) smoked. Of these 744 people, 123 (16%) had counseling and 651 did not. Age averaged 50 years in counseled smokers and 47 in uncounseled smokers (P = 0.01). Respective proportions of women were 47% and 43%, blacks 85% and 85%, and uninsured 29% and 40% (P = 0.03). Among all patients analyzed, 52% had a history of mental illness, 58% had used cocaine, and 41% had used heroin. The study group smoked 11 cigarettes daily and had tried to quit an average of 2 times.
Among the 123 counseled smokers, 101 (82%) got prescribed a smoke-ending drug, usually nicotine replacement therapy (96% of 101), though 54% got a combination prescription. Most people with a prescription (83%) reported nonadherence at least once during follow-up, and 48% said insurance did not fully cover their prescription.
Twenty of 123 people (16%) who received counseling reported quitting at 6 months. (The study did not confirm quitting biochemically.) Multivariate analysis determined that past or current heroin or cocaine use lowered chances of quitting 80% (aOR 0.20, 95% CI 0.07 to 0.56, P = 0.002). Being in the preparation stage to quit rather than the precontemplation stage or the contemplation stage boosted odds of quitting more than 8 times (aOR 8.26, 95% CI 1.02 to 66.67, P = 0.048). Factors that did not influence quitting in this analysis were age, gender, race, mental illness history, and nicotine dependence level.
Pluses and Minuses of Four Nondrug Interventions
Each of the four just-reviewed nondrug interventions has its pluses and minuses (Table 3). PositivelySmokeFreeMe is free and immediately available to all smokers with Internet access.39 This 8-session Web-based program may be particularly well-suited to people who enjoy interactive online programs. But unless clinicians or staff can check patient progress through the eight online segments, many smokers may not have the motivation needed to complete the program. The inner-city study group in a randomized trial39 required regular prodding to access each new session that came online. So PositivelySmokeFreeMe will probably work best for more highly motivated individuals. Clinicians will have to work with motivated patients to select and prescribe an appropriate medication and to promote adherence.
Table 3. Four Nondrug Interventions* Tested in People With HIV
Trial, Smoking Level
8-Session Internet program for smokers39
Bronx, NY, n = 138,† age 46, 40% women, 75% black, 47% Hispanic, 86% unemployed
Randomized, 11 cig/day
Ready-to-use, free online program developed for HIV+ smokers
Text reminders and calls needed to ensure adherence
10% at 6 m, up to 40% with good adh
Cell-phone reminder calls40
Houston, n = 474,† age 43-46, 30% women, 75% black, 9% Hispanic, 80% unemployed
Randomized, 19 cig/day
Cell-phone use simplifies counseling, follow-up
Staff time required for regular cell-phone follow-up
12%-15% at 3 m
Half-day clinician training41
Zurich SHCS, n = 1689, age 44, 25% women, 22% IDU history
Prospective cohort, 20 pack-years
Ensures provider training in half-day session; uses simple patient checklist
Assumes provider adherence (80% in this study)
23% higher odds of quitting with provider training
One-time 1-hour, 1-on-1 smoker counseling43
Newark, NJ, n = 123,† age 50, 47% women, 85% black, 52% mental illness history, 58% cocaine history
Prospective cohort, 11 cig/day
Time efficient, focused on individual smoker, can be done by trained nonphysicians, including patient peers
Ensures only 1 counseling session, therapy may have to be prescribed at separate visit
16% at 6 m
* All interventions coupled with drug therapy prescription, offer, or advice.
† All participants said they wanted to quit smoking.
adh, adherence; cig, cigarettes; IDU, injection drug use; SHCS, Swiss HIV Cohort Study.
Cell-phone counseling calls offer a reliable platform to bolster patient commitment to quitting, to remind patients to use their antismoking medication, and to address concerns that arise.40 But such a program requires some planning at the clinic level, commitment of time from capable staff or perhaps trained peer volunteers, and occasional interaction with the clinician. Houston researchers who tested this approach found that quit rates waned when the calls stopped.
The half-day clinician training program designed by the Zurich unit of the Swiss HIV Cohort Study (SHCS) provides the foundation for any successful smoking cessation program by ensuring that providers understand the principles of nicotine addiction, counseling, and treatment.41 After the Zurich researchers launched this program, which included a smoking checklist for each patient, they recorded higher quit rates and lower relapse rates than other SHCS units across Switzerland. Program success clearly depends on clinicians implementing what they learn in the session, as a large majority of clinicians in the Zurich study did. After publishing their report, the Zurich group began providing a 1-hour smoking cessation course for HIV clinicians once a year. Other SHCS centers have not adopted such a program. Updated numbers show that smoking prevalence in HIV-positive people remains lower in Zurich than at other centers.42
The primary advantage of the one-time, 1-hour, one-on-one smoke-ending session is its focus on individual smokers in a single session that can be performed by a nonphysician health professional or even a peer counselor.43 The Newark researchers who ran this study observe that this "low-intensity" approach yielded a 6-month quit rate similar to rates recorded in studies of more complicated and lengthy interventions. Yet the authors note that prior or current illicit drug use cut chances of success, a finding suggesting some smokers "may benefit from more intensive cessation approaches or strategies that incorporate substance use counseling and mental health services into cessation interventions."43
The three US studies all involved low-income heavy smokers with high rates of alcohol and drug abuse and psychiatric comorbidities.39,40,43 All achieved similar short-term quit rates around 15%. Although that rate may sound low, it is comparable to rates in some general population studies, which often include people with fewer addiction and dependence risk factors. And getting 3 of every 20 HIV-positive people to quit will certainly have a profound impact on the health of those quitters -- an impact that easily justifies the health worker effort.
* Quit ratio equals former smokers divided by former plus current smokers.
National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365:395-409.
Park ER, Gareen IF, Japuntich S, et al. Primary care provider-delivered smoking cessation interventions and smoking cessation among participants in the National Lung Screening Trial. JAMA Intern Med. 2015;175:1509-1516.
Crothers K, Goulet JL, Rodriguez-Barradas C, et al. Decreased awareness of current smoking among health care providers of HIV-positive compared to HIV-negative veterans. J Gen Intern Med. 2007;22:749-754.
Jha P, Ramasundarahettige C, Landsman V, et al. 21st-century hazards of smoking and benefits of cessation in the United States. N Engl J Med. 2013;368:341-350.
Lifson AR, Neuhaus J, Arribas JR, et al. Smoking-related health risks among persons with HIV in the Strategies for Management of Antiretroviral Therapy Clinical Trial. Am J Public Health. 2010;100:1896-1903.
Petoumenos K, Worm S, Reiss P, et al. Rates of cardiovascular disease following smoking cessation in patients with HIV infection: results from the D:A:D study. HIV Med. 2011;12:412-421.
De P, Farley A, Lindson N, Aveyard P. Systematic review and meta-analysis: influence of smoking cessation on incidence of pneumonia in HIV. BMC Med. 2013;11:15.
Gordin FM, Roediger MP, Girard PM, et al. Pneumonia in HIV-infected persons: increased risk with cigarette smoking and treatment interruption. Am J Respir Crit Care Med. 2008;178:630-636.
Bénard A, Mercié P, Alioum A, et al. Bacterial pneumonia among HIV-infected patients: decreased risk after tobacco smoking cessation. ANRS CO3 Aquitaine Cohort, 2000-2007. PLoS One. 2010;5:e8896.
Calvo-Sánchez M, Martínez E. How to address smoking cessation in HIV patients. HIV Med. 2015;16:201-210.
Shuter J, Salmo LN, Shuter AD, Nivasch EC, Fazzari M, Moadel AB. Provider beliefs and practices relating to tobacco use in patients living with HIV/AIDS: a national survey. AIDS Behav. 2012;16:288-294.
Tornero C, Mafé C. Varenicline and antiretroviral therapy in patients with HIV. J Acquir Immune Defic Syndr. 2009;52:656.
Cui Q, Robinson L, Elston D, et al. Safety and tolerability of varenicline tartrate (Champix/Chantix) for smoking cessation in HIV-infected subjects: a pilot open-label study. AIDS Patient Care STDS. 2012;26:12-19.
Ferketich AK, Diaz P, Browning KK, et al. Safety of varenicline among smokers enrolled in the lung HIV study. Nicotine Tob Res. 2013;15:247-254.
US National Library of Medicine. MedlinePlus. Bupropion.
Shirley DK, Kaner RJ, Glesby MJ. Effects of smoking on non-AIDS-related morbidity in HIV-infected patients. Clin Infect Dis. 2013;57:275-282.
Pedrol-Clotet E, Deig-Comerma E, Ribell-Bachs M, Vidal-Castell I, García-Rodríguez P, Soler A. Bupropion use for smoking cessation in HIV-infected patients receiving antiretroviral therapy. Enferm Infecc Microbiol Clin. 2006;24:509-511.
Lloyd-Richardson EE, Stanton CA, Papandonatos GD, et al. Motivation and patch treatment for HIV+ smokers: a randomized controlled trial. Addiction. 2009;104:1891-1900.
Humfleet GL, Hall SM, Delucchi KL, Dilley JW. A randomized clinical trial of smoking cessation treatments provided in HIV clinical care settings. Nicotine Tob Res. 2013;15:1436-1445.
Martín Cantera C, Puigdomènech E, Ballvé JL, et al. Effectiveness of multicomponent interventions in primary healthcare settings to promote continuous smoking cessation in adults: a systematic review. BMJ Open. 2015;5:e008807.
Fuster M, Estrada V, Fernandez-Pinilla MC, et al. Smoking cessation in HIV patients: rate of success and associated factors. HIV Medicine. 2009;10:614-619.
Cropsey KL, Jardin BF, Burkholder GA, Clark CB, Raper JL, Saag MS. An algorithm approach to determining smoking cessation treatment for persons living with HIV/AIDS: results of a pilot trial. J Acquir Immune Defic Syndr. 2015;69:291-298.
Bullen C, Howe C, Laugesen M, et al. Electronic cigarettes for smoking cessation: a randomised controlled trial. Lancet. 2013;382:1629-1637.
Farsalinos KE, Romagna G, Tsiapras D, Kyrzopoulos S, Voudris V. Evaluating nicotine levels selection and patterns of electronic cigarette use in a group of "vapers" who had achieved complete substitution of smoking. _Subst Abus_e. 2013;7:139-146.
US Food and Drug Administration. E-cigarettes: questions and answers. August 30, 2015.
Shuter J, Morales DA, Considine-Dunn SE, An LC, Stanton CA. Feasibility and preliminary efficacy of a web-based smoking cessation intervention for HIV-infected smokers: a randomized controlled trial. J Acquir Immune Defic Syndr. 2014;67:59-66.
Gritz ER, Danysh HE, Fletcher FE, et al. Long-term outcomes of a cell phone-delivered intervention for smokers living with HIV/AIDS. Clin Infect Dis. 2013;57:608-615.
Huber M, Ledergerber B, Sauter R, et al. Outcome of smoking cessation counselling of HIV-positive persons by HIV care physicians. HIV Medicine. 2012;13:387-397.
Bruno Ledergerber. University of Zurich. Email November 13, 2015.