Secondary Logo

Journal Logo

Clinical Science

An Algorithm Approach to Determining Smoking Cessation Treatment for Persons Living With HIV/AIDS

Results of a Pilot Trial

Cropsey, Karen L. PsyD*; Jardin, Bianca F. PhD; Burkholder, Greer A. MD; Clark, C. Brendan PhD*; Raper, James L. PhD, JD; Saag, Michael S. MD

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: July 1, 2015 - Volume 69 - Issue 3 - p 291-298
doi: 10.1097/QAI.0000000000000579



People living with HIV (PLHIV) are currently losing more years of life because of smoking than to their HIV disease.1 Continued smoking among PLHIV has unique and specific risks that promote more rapid development of cardiovascular disease, pulmonary disease, bacterial infections, malignancy, and other adverse health outcomes.2–4 It is now estimated that smoking is responsible for over 60% of deaths among PLHIV.1 The health implications of continued smoking in this population are widespread, because approximately 50%–70% of PLHIV are current smokers, a rate that is approximately 3 times higher than the general population.5–7 As a result, smoking now represents the most significant modifiable risk factor for disease and mortality in PLHIV.

Although a broad spectrum of effective behavioral and pharmacologic services are available to help these smokers quit, utilization of these evidence-based cessation treatments remains extremely low.8 Utilization rates may be related to the reactive approach of our current health care model to smoking cessation. Specifically, current clinical practice relies on smokers to either request treatment services or depends on the provider to initiate a conversation about smoking cessation.9 Unfortunately, because less than 10% of smokers are ready to quit in the immediate future, a reactive approach may result in significant missed opportunities to engage smokers in cessation efforts.10 More recent trends in treatment-based studies have focused on expanding the reach of cessation efforts by proactively identifying smokers and offering intervention efforts.9 Proactive components have included identifying smokers through clinical registries or online/published channels and providing mailed outreach of cessation aids, telephone counseling, etc.9–11 In studies with PLHIV, proactive approaches have shown good success, even when targeting unmotivated smokers.12–14

The HIV/AIDS clinical setting remains a prominent point of health care contact for many PLHIV smokers. However, the successful integration of proactive models of care in this clinical setting will require that any cessation service offered dovetail easily with existing HIV/AIDS treatment regimens. Currently, very few HIV providers report receiving tobacco cessation treatment training and most report little time or confidence in their ability to assess and treat tobacco dependence during a routine clinical encounter.15 Algorithm-based interventions have been shown to be an effective clinical tool that can be used in busy clinical settings because they concisely synthesize knowledge and provide greater guidance with treatment decisions by more clearly prioritizing which factors need to be considered when prescribing pharmacotherapy.16–19 Algorithms also have the added advantage of allowing for greater flexibility in response to evolving patient needs.17,18 Although algorithms have been successfully developed to standardize and provide guidance for best practices in the treatment of many disorders,20,21 they have rarely been integrated into the field of smoking cessation.

In the current randomized pilot clinical trial, we tested a proactive model of smoking cessation care versus usual care. We used clinical registry information to identify smoking status, and through the assistance of an algorithm, proactively offered these smokers different forms of pharmacotherapy regardless of baseline motivation to quit or not. The main purpose of this study was to determine if proactively identified smokers who received algorithm-directed intervention at their HIV appointment would alter behaviors associated with smoking cessation compared with participants who received Treatment as Usual (TAU).


Participant Recruitment and Eligibility

One hundred and thirty four participants were approached to participate in this study, and 102 participants provided informed consent (for study flow, see Figure S1, Supplemental Digital Content, Two participants were excluded after consent; one was not receiving medical care at the clinic and the other participant tested negative for cotinine and was not smoking. Participants were enrolled in the study if they were at least 19 years of age (the legal age of adulthood in Alabama), receiving their HIV care at the clinic and not anticipating moving or changing clinics over the next 6 months, smoking at least 5 cigarettes per day for the past month, tested positive for cotinine, and living in an unrestricted environment that allowed smoking. Participants were excluded if they were pregnant or nursing, not English-speaking, or had cognitive impairment such that they could not provide informed consent.

All patients at the University of Alabama at Birmingham 1917 HIV clinic completed patient-reported outcomes (PROs) every 6 months as part of routine clinical care and included in these measures was a question that determined smoking status. We identified potential participants through their most recent responses on the PROs. Participants who indicated current smoking during their last clinic visit were approached by their medical provider to determine if they were interested in being in this study. Research staff were available in the clinic 2 days per week and enrolled all consenting participants during their clinic visit. Participants who were not able to be enrolled during their clinic visit were scheduled for a baseline visit during the next week.

A total of 100 participants were deemed eligible to participate in the study and were subsequently randomized to receive either algorithm-derived treatment (AT) for smoking cessation or TAU, which was smoking cessation assistance from their medical provider when the patient was ready to quit.


All enrolled participants completed baseline assessments that included demographic, smoking history, and psychological constructs such as motivation and abstinence self-efficacy. Participants also completed the Fagerström Test of Cigarette Dependence22 and the Minnesota Nicotine Withdrawal Scale,23 which assessed dependence on cigarettes and nicotine withdrawal, respectively. The Thoughts About Abstinence questionnaire (TAA24) was administered to assess motivation, abstinence self-efficacy, perceived difficulty of quitting, and goal for smoking. Goal for smoking was assessed using a multiple-choice answer option ranging from complete lifelong abstinence to reduction in smoking to no change in smoking behavior. All measures were self-reported surveys completed by participants during their visit.

These measures were combined with other measures of interest in the patient's PRO. This included a measure of depressive symptoms [Patient Health Questionnaire-Depression (PHQ-9D)], panic symptoms [Patient Health Questionnaire-Anxiety (PHQ-A)], tobacco, alcohol and other drug use [Alcohol Smoking, and Substance Involvement Screening Test (WHO ASSIST v3.0)], and hazardous alcohol use [3-item Alcohol Use Disorders Identification Test (AUDIT-C)]. Nadir CD4 count, viral load, and chronic medical conditions were extracted from the clinical database for all participants and smokers who were seen in clinic but did not enroll in the study. In addition to completing demographic information, during the baseline visit, regardless of treatment group or abstinence goal, all participants received 1 standard 20-minute smoking cessation counseling at baseline provided by the research staff who discussed behavioral strategies for cutting down or reducing smoking. Research staff had a bachelor's level education and did not possess any specific training in smoking cessation other than a brief training to review common strategies used for smoking cessation used in this intervention. The strategies that were covered during the sessions included reviewing the importance of quitting, strategies to avoid or cope with high-risk situations, and preparing to quit including telling friends and family their intention to quit and removing smoking paraphernalia and tobacco products from the home. The importance of using pharmacotherapy to assist with cessation was discussed with all participants.

After baseline assessments, research staff asked questions according to the algorithm to determine pharmacotherapy selection (see Fig. 1 for a schematic of the algorithm). All participants were initially asked if they were interested in quitting smoking today or in the next 30 days. If the participant said “yes,” they were then asked if they were willing to take a medication that they would have to take by mouth twice per day. Because varenicline has been shown to have superior quit rates over other single pharmacotherapy agents, it was selected as the first line of treatment for patients willing to take a medication twice per day and wanting cessation.25 If varenicline was contraindicated (eg, severe renal impairment), bupropion was the second agent for patients who were interested in quitting and willing to take a medication. Single nicotine replacement therapy (NRT) was the third agent if varenicline or bupropion were not appropriate, with nicotine patch as the first line NRT offered. In preference for NRT, lozenge was offered as the second NRT option, gum was the third, inhaler was the fourth, and nicotine nasal spray was the last option for NRT. The order of NRT within the algorithm was based on patient familiarity and number of clinical trials supporting their use.26 Clinical trials have also supported the use of combination NRT (eg, patch and lozenge),27 and this was offered to participants who made an unsuccessful quit attempt with NRT in the past. Similarly, among participants not interested in quitting, they too were offered NRT, if no contraindications existed. Participants were told they could use the NRT to practice quitting, in situations in which they were not allowed to smoke, or to reduce their smoking. The delivery of NRT to this sample was predicated on recent findings that NRT can serve as an external cue to positively shift quitting momentum and ultimately trigger action.10,11 After a period of 4 weeks, participants who continued to smoke and were initially prescribed a single agent were then offered an additional pharmacotherapy agent, generally nicotine lozenge for acute nicotine cravings, to aid in cessation. Although the algorithm questions were asked by study staff, the clinic provider approved all medications for participants.

Algorithmic treatment of smoking cessation.

After the baseline, participants entered a 12-week follow-up period, during which in-person assessments were made at weeks 2, 4, 8, and 12 (the end of treatment). Final follow-up was at 1 month posttreatment. During each of the assessment periods, study staff collected data about several primary measures of interest, which included reduction in smoking behavior as measured by the number of cigarettes smoked per day (CPD) and any 24-hour quit attempt, as per Centers for Disease Control and Prevention definition.28 We also collected secondary measures of interest focused on process measures of cessation including abstinence self-efficacy, motivation to quit, perceived difficulty of quitting; the latter constructs were assessed using a single-item 0-10 measure and were derived from the TAA questionnaire. Additionally, at each assessment period, we assessed uptake of pharmacotherapy resources, as provided by us or purchased independently [yes/no].

Statistical Analyses

Chi-square and analysis of variance were used to determine equivalence between groups on baseline characteristics. Generalized estimating equations (GEE) models were used to determine the effect of the intervention across study time points for both our primary measures of interest, ie, changes in smoking behaviors and quit attempts, and the process measure of cessation including self-efficacy to quit, motivation to quit, perceived difficulty of quitting, and uptake of pharmacotherapy resources. Main effects of these variables and interaction terms were examined across time. An intent-to-treat analysis was conducted, and all missing values were coded as smoking.


Baseline Comparisons Between Algorithm and Treatment as Usual

Tables 1 and 2 presents a summary of baseline characteristics for both groups. Overall, no differences between groups were detected for demographic, psychosocial, or smoking history variables. About two-thirds reported a goal of lifelong abstinence from smoking.

Demographic Characteristics (N = 100)
Baseline Smoking Characteristics (N = 100)

Medication Use and Adherence

Participants randomized to algorithm treatment most often received nicotine patch (38%), followed by varenicline (36%), combination NRT (nicotine patch and lozenge; 12%), bupropion (10%), or lozenge (4%). At week 4, over half of participants (52%) were offered a second pharmacotherapy agent because they had not made a quit attempt and were still smoking; specifically, they were given nicotine lozenges to cope with acute cravings. Thus, 89% of participants receiving nicotine patches added lozenge, 40% of bupropion smokers added lozenges, and 39% of participants who received varenicline added lozenge. Overall, slightly over half (57.1%) of the participants who received medication took the medication as prescribed.

Smoking Outcomes

As shown in Supplemental Table S1 (see Supplemental Digital Content,, a significant interaction between treatments by time was found for number of CPD such that smokers who received AT reduced the number of CPD across time relative to the TAU (AT: 10 CPD reduction vs. TAU: 6 CPD from baseline to 1-month follow-up; see Fig. 2). Similarly, across the follow-up period, the proportion of participants reporting 24-hour quit attempts was significant higher for participants who received AT compared with TAU participants (50% vs. 38%). Finally, more participants who received AT used medication for smoking cessation compared with TAU (81% vs. 23% overall; see Table S1, Supplemental Digital Content, GEE analyses showed a significant interaction for treatment across time for medication utilization such that AT and TAU reported similar past use of smoking cessation pharmacotherapies at baseline; however, AT participants reported increased utilization of medications while TAU reported a decrease in use of medications across time (see Fig. 3).

Interaction between treatments across time for average number of cigarettes smoked per day (n = 100).
Interaction between treatments across time for medication utilization.

Give the high percentage of participants who reported use of other tobacco products, particularly cigars and cigarillos, we examined the use of any other tobacco product during the past week across time by treatment group using GEE analysis. We found no significant interaction between time and treatment, and no main effect of treatment. However, there was a significant main effect of time such that during the intervention period, use of these products increased for both groups, with 7% reporting use of another tobacco product (primarily cigars/cigarillos) at baseline and increasing to 17% at weeks 4 and 8 and dropping back to 6% by week 16 (Wald χ2 = 14.7; P = 0.012).

Measures of Cessation Readiness

GEE models' testing changes motivation, self-efficacy, and perceptions of quitting difficulty and revealed significant main effects for treatment group for all constructs such that participants assigned to AT group reported higher levels of motivation (M = 8.93 vs. 8.61; P = 0.015) and abstinence self-efficacy (M = 8.69 vs. 8.14; P < 0.001), and lower levels of perceptions of quitting difficulty (M = 4.65 vs. 5.46; P < 0.001) as compared with participants assigned to TAU over time (see Table S2, Supplemental Digital Content,

Comparison of Sample to Clinic Population

Given the largely proactive nature of recruitment for this study, we wanted to ensure that participants recruited in our study were representative of the larger population of PLHIV being treated in the clinic. Specifically, we compared individuals who participated in this study to smokers who were seen in the clinic during the recruitment time but who were not approached to be in the study (n = 100, n = 356; see Table S3, Supplemental Digital Content, Study participants were older at the time of the study and older when they first were diagnosed with HIV. Female and African American participants were overrepresented in the intervention relative to smokers in the HIV clinic. Study participants had a higher Nadir CD4 count compared with smokers in the clinic. Finally, participants in the intervention were more likely to be currently using cocaine but were less likely to engage in risky alcohol use compared with clinic smokers. No differences were found between study and nonstudy participants on all other PROs, and no significant differences were found among individuals in the algorithm group compared with TAU participants on any of these variables.


This is the first randomized pilot clinical trial to examine the implementation and delivery of a novel proactive algorithm-based intervention among PLHIV smokers engaged in HIV medical care. The findings demonstrate that our intervention was well integrated into the clinic setting and that the provision of active treatments to smokers receiving treatment in an HIV setting can produce important clinical changes, including reduced cigarette consumption, elevated levels of abstinence self-efficacy and motivation, and diminished perception of difficulty for quitting, compared with participants receiving usual care. Importantly, these findings were inclusive of both motivated and unmotivated smokers. The overall evidence from this pilot is promising and has strong potential for dissemination into clinical settings.

Smoking is increasingly becoming the most important modifiable risk factor for morbidity and mortality among PLHIV, and therefore it is unequivocal that smoking cessation becomes a top priority for HIV care providers. Unfortunately, the complexity of HIV management often renders smoking cessation a lower priority among providers, who otherwise lack the training and confidence to initiate cessation initiatives, particularly counseling-based cessation services.29 Although counseling services have significant merit, counseling is also often time-intensive and not compatible with a busy and often overburdened HIV clinical setting. The algorithm protocol presented herein offers an alternative method for incorporating smoking cessation efforts more easily into clinical encounters.19 Specifically, this easy-to-follow tool provides clinicians with a brief and succinct pathway for treatment selection decisions and thereby lends itself to being used during even brief clinical encounters.19 In addition to being time-efficient, this tool capitalizes on other components of optimal health care including patient customization.17 The latter is particularly important for an HIV population, many of whom face unique barriers to smoking cessation given high rates of comorbid drug and alcohol use and mental illness.30 These factors not only make cessation more challenging but also render some cessation resources inappropriate, eg, bupropion for patients with a history of bipolar disorder.31

Although the overall pattern of results suggests that the algorithm protocol is both feasible and produced encouraging results, this study has a number of limitations inherent in the design. As a pilot study, our sample was small, and therefore we had insufficient power to detect more desired clinical outcomes, eg, prolonged abstinence. Nonetheless, we did focus on other well-established process measures of cessation,32 which we believe to be an important, and cost-efficient, first step for studies focused on initially establishing feasibility.33 In light of our promising results, however, the next steps should include randomized clinical trials using larger samples and targeting more rigorous clinical outcomes over a greater period of time. In addition, because of the pilot nature of this project and limited funding, research staff could only be in the clinic during 2 days for active recruitment and were not available to approach all eligible participants. However, our study sample was fairly similar to the clinic sample, although females and African American smokers were overrepresented relative to rates in the clinic. However, this is likely the strength of our study, particularly given the general lack of representation of minority patient populations in current HIV treatment studies.34 Sociodemographic differences in the prevalence of HIV/AIDS are well established, often with African Americans bearing the greatest burden and demonstrating differences in treatment responsiveness. Finally, some participants were unable to be seen after their regularly scheduled appointment and scheduled their appointment outside this time to do the study. Unfortunately, this was not tracked; therefore, it is impossible to determine how this may have impacted our results.

In conclusion, this trial provides evidence that a proactive algorithm-based protocol was effective in producing changes across a number of important clinical markers associated with smoking cessation.32 In order for a smoking intervention to produce meaningful changes in population-level cessation rates, it must not only demonstrate potential for efficacy but also be able to reach a large proportion of smokers.35 Because the capacity to identify smokers more easily through electronic medical records continues to grow, the use of proactive initiatives will represent an excellent approach in which to increase smokers' access to evidence-based cessation resources.9 Beyond improvements in reach, proactive initiatives circumvent several barriers associated with physician-delivered smoking cessation, including lack of time and interest, biases about smokers' motivation, and racial/ethnic background.9 Combined, these advantages suggest strong potential to further advance clinical health outcomes in this population.


The authors thank Ms. Lei Li and Mr. Nandan Katiyar for their assistance with data collection.


1. Helleberg M, Afzal S, Kronborg G, et al.. Mortality attributable to smoking among HIV-1-infected individuals: a nationwide, population-based cohort study. Clin Infect Dis. 2013;56:727–734.
2. Rahmanian S, Wewers ME, Koletar S, et al.. Cigarette smoking in the HIV-infected population. Proc Am Thorac Soc. 2011;8:313–319.
3. Mangili A, Polak JF, Skinner SC, et al.. HIV infection and progression of carotid and coronary atherosclerosis: the CARE study. J Acquir Immune Defic Syndr. 2011;58:148–153.
4. De P, Farley A, Lindson N, et al.. Systematic review and meta-analysis: influence of smoking cessation on incidence of pneumonia in HIV. BMC Med. 2013;11:15.
5. Niaura R, Shadel WG, Morrow K, et al.. Human immunodeficiency virus infection, AIDS, and smoking cessation: the time is now. Clin Infect Dis. 2000;31:808–812.
6. Benard A, Bonnet F, Tessier JF, et al.. Tobacco addiction and HIV infection: toward the implementation of cessation programs. ANRS CO3 Aquitaine Cohort. AIDS Patient Care and STDS. 2007;21:458–468.
7. 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.
8. Fiore C, Jaen CR. Treating Tobacco Use and Depencence: 2008 Update. Clinical Practice Guideline. Rockville, MD: US Public Health Service; 2008.
9. Fu SS, van Ryn M, Sherman SE, et al.. Proactive tobacco treatment and population-level cessation: a pragmatic randomized clinical trial. JAMA Intern Med. 2014;174:671–677.
10. Jardin BF, Cropsey KL, Wahlquist AE, et al.. Evaluating the effect of access to free medication to quit smoking: a clinical trial testing the role of motivation. Nicotine Tob Res. 2014;16:992–999.
11. Carpenter MJ, Hughes JR, Gray KM, et al.. Nicotine therapy sampling to induce quit attempts among smokers unmotivated to quit: a randomized clinical trial. Arch Intern Med. 2011;171:1901–1907.
12. Vidrine DJ, Arduino RC, Lazev AB, et al.. A randomized trial of a proactive cellular telephone intervention for smokers living with HIV/AIDS. AIDS. 2006;20:253–260.
13. Vidrine DJ, Marks RM, Arduino RC, et al.. Efficacy of cell phone-delivered smoking cessation counseling for persons living with HIV/AIDS: 3-month outcomes. Nicotine Tob Res. 2012;14:106–110.
14. Cropsey KL, Hendricks PS, Jardin B, et al.. A pilot study of screening, brief intervention, and referral for treatment (SBIRT) in non-treatment seeking smokers with HIV. Addict Behav. 2013;38:2541–2546.
15. Shuter J, Bernstein SL, Moadel AB. Cigarette smoking behaviors and beliefs in persons living with HIV/AIDS. Am J Health Behav. 2012;36:75–85.
16. Bader P, McDonald P, Selby P. An algorithm for tailoring pharmacotherapy for smoking cessation: results from a Delphi panel of international experts. Tob Control. 2009;18:34–42.
17. Hughes J. An algorithm for choosing among smoking cessation treatments. J Subst Abuse Treat. 2008;34:426–432.
18. Hughes JR. An updated algorithm for choosing among smoking cessation treatments. J Subst Abuse Treat. 2013;45:215–221.
19. Jones SJ, Gardner CL, Cleveland KK. Development of a smoking cessation algorithm for primary care providers. J Nurse Pract. 2014;10:120–127.
20. Barst RJ, Gibbs JS, Ghofrani HA, et al.. Updated evidence-based treatment algorithm in pulmonary arterial hypertension. J Am Coll Cardiol. 2009;54(1 suppl):S78–S84.
21. Nathan DM, Buse JB, Davidson MB, et al.. Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2009;32:193–203.
22. Fagerström K. Determinants of tobacco use and renaming the FTND to the Fagerstrom test for cigarette dependence. Nicotine Tob Res. 2012;14:75–78.
23. Hughes JR, Hatsukami D. Signs and symptoms of tobacco withdrawal. Arch Gen Psychiatry. 1986;43:289–294.
24. Hendricks PS, Wood SB, Baker MR, et al.. The Smoking Abstinence Questionnaire: measurement of smokers' abstinence-related expectancies. Addiction. 2011;106:716–728.
25. Garrison GD, Dugan SE. Varenicline: a first-line treatment option for smoking cessation. Clin Ther. 2009;31:463–491.
26. Stead LF, Perera R, Bullen C, et al.. Nicotine replacement therapy for smoking cessation. Cochrane Database Syst Rev. 2012;11:CD000146.
27. Carpenter MJ, Jardin BF, Burris JL, et al.. Clinical strategies to enhance the efficacy of nicotine replacement therapy for smoking cessation: a review of the literature. Drugs. 2013;73:407–426.
28. USDHS. Cigarette smoking–United States, 1965-2005. MMWR. 2011;60:109–113.
29. Moscou-Jackson G, Commodore-Mensah Y, Farley J, et al.. Smoking-cessation interventions in people living with HIV infection: a systematic review. J Assoc Nurses AIDS Care. 2014;25:32–45.
30. Bing EG, Burnam MA, Longshore D, et al.. Psychiatric disorders and drug use among human immunodeficiency virus-infected adults in the United States. Arch Gen Psychiatry. 2001;58:721–728.
31. Dervaux A, Laqueille X. Antidepressant treatment and smoking cessation in bipolar disorder. JAMA. 2009;301:2093; author reply.
32. Vangeli E, Stapleton J, Smit ES, et al.. Predictors of attempts to stop smoking and their success in adult general population samples: a systematic review. Addiction. 2011;106:2110–2121.
33. Carpenter MJ, Alberg AJ, Gray KM, et al.. Motivating the unmotivated for health behavior change: a randomized trial of cessation induction for smokers. Clin Trials. 2010;7:157–166.
34. Cioe PA. Smoking cessation interventions in HIV-infected adults in North America: a literature review. J Addict Behav Ther Rehabil. 2013;2:1000112.
35. Zhu SH, Lee M, Zhuang YL, et al.. Interventions to increase smoking cessation at the population level: how much progress has been made in the last two decades? Tob Control. 2012;21:110–118.

HIV; AIDS; smoking cessation; algorithm; proactive

Supplemental Digital Content

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.