Cigarette Smoking Behaviors and Beliefs in Persons Living With Hepatitis C

Abstract

Background and Rationale:

Tobacco use is common among persons living with hepatitis C (PLHC), yet little is known about their smoking behaviors and beliefs. Modern hepatitis C treatment offers a unique opportunity to intensively engage this population about other health risks, including smoking.

Main Results:

Seventy-seven tobacco users (40 hepatitis C virus [HCV] seropositive and 37 HCV seronegative) enrolled in an interview study in a New York City clinic. The mean age was 51.6, 57.1% were male, 40.3% Latino, and 49.4% black. 67.5% were single and 18.2% were employed. HCV+ smokers differed from HCV− smokers in having a higher prevalence of illicit substance use, depression, and hypertension. PLHC smokers were highly motivated to quit, with 52.5% stating an intention to quit within 30 days. Most of the PLHC smokers had used cessation-directed pharmacotherapy, but almost none had tried a quitline or a quit smoking website. PLHC smokers scored higher on the intrapersonal locus of control subscale. Almost a quarter (22.5%) believed that smoking “helped fight the HCV.”

Conclusions:

PLHC smokers have a high burden of psychiatric and substance use comorbidity. They exhibit characteristics that distinguish them from uninfected smokers, and many harbor false beliefs about imagined benefits of smoking. They are highly motivated to quit but underutilize cessation aids. Without aggressive intervention, smoking-related morbidity will likely mute the health benefits and longevity gains associated with hepatitis C treatment. Research such as this may prove useful in guiding the development of future tobacco treatment strategies.

Implications:

This is the first paper to examine, in detail, sociobehavioral correlates of tobacco use in PLHC. PLHC are recognized by the Department of Health and Human Services as a high-priority health disparities population. We are not aware of any tobacco treatment services designed specifically for PLHC. The first step in designing an intervention is defining the characteristics of the target group. Our findings will begin to address this need, and may prove useful in optimizing tobacco treatment strategies for smokers living with hepatitis C.

Introduction

Forty million US residents are current cigarette smokers,¹ and close to 3 million Americans live with chronic hepatitis C infection.² Although the hepatitis C epidemic in the United States is concentrated in groups with high rates of smoking, such as African Americans and people living below the poverty level,^1,2 there are few population based statistics on smoking rates in persons living with hepatitis C (PLHC). A recent, national Veteran’s Administration survey demonstrated a current smoking prevalence of 67% in a sample of more than 111 000 hepatitis C-infected individuals enrolled in the Veteran’s Administration system.³ In the 2011–2012 National Health and Nutrition Examination Survey, 56.8% of hepatitis C seropositive individuals were current tobacco users as compared to 20.6% of hepatitis C seronegative respondents (odds ratio [OR] = 5.06 [3.30–7.75]).⁴ These data suggest that hepatitis C-infected individuals in the United States smoke at approximately three times the rate of the general population.

Substance use and psychiatric illness are overrepresented in PLHC.⁵ These comorbidities complicate tobacco treatment and may necessitate specialized and more intensive therapeutic strategies.^6,7 Moreover, tobacco use is linked to multiple adverse outcomes in PLHC. There is evidence to suggest that cigarette smoking is associated with intrahepatic complications of hepatitis C, including steatosis, fibrosis, and hepatocellular carcinoma.^8,9 Tobacco use is also associated with extrahepatic complications in PLHC, including atherosclerotic disease,¹⁰ arthralgias,¹¹ lower quality of life,¹² and increased likelihood of delayed or deferred hepatitis C treatment.¹³ Whereas past hepatitis C virus [HCV] therapies have been relatively ineffective, new oral treatment regimens with direct-acting antivirals are substantially more efficacious, with few side effects, short treatment durations, and cure rates as high as 98%.¹⁴ Importantly, successful HCV treatment has been associated with decreased mortality.¹⁵ As more patients are cured of HCV, smoking-related morbidity and mortality are likely to increase in importance, undermining the health outcomes achieved by the advances in HCV treatment.

Despite the importance of tobacco use in this population, we are not aware of any published studies characterizing the sociobehavioral correlates of cigarette smoking behaviors of PLHC. Increasing numbers of PLHC are accessing the medical establishment in order to obtain new, direct acting-antiviral treatments. The treatment period includes frequent and intense contact with medical providers over a course of at least 8 weeks. The treatment process and these frequent contacts could serve as teachable moments during which providers may impact health outside the sphere of hepatitis C disease. Conversely, each failure to engage PLHC entering treatment represents a missed opportunity to promote smoking cessation and other healthy life choices.

We are not aware of any tobacco treatments specifically targeting the PLHC population. The first step in developing appropriately targeted interventions is acquisition of an in-depth understanding of the biopsychosocial correlates and drivers of tobacco use in PLHC and how these variables differ in PLHC compared to other smokers. Our study hypothesis was that hepatitis C-infected smokers would differ in important ways, especially relating to other substance use and psychiatric illness, from hepatitis C-uninfected smokers receiving care in the same clinical setting. Uncovering these differences may prove useful in optimizing tobacco treatment strategies for this highly vulnerable group. In order to test the hypothesis, we completed interviewer-administered questionnaires on a sample of hepatitis C-infected and uninfected smokers in a single clinic in the Bronx, New York.

Methods

Participants were recruited for the study from the patient population of Montefiore Medical Center’s Comprehensive Health Care Center. Comprehensive Health Care Center is a federally qualified health center that offers comprehensive medical care to the population living in surrounding areas of the South Bronx, a low-income neighborhood that is 57% Hispanic and 39% non-Hispanic black. Services provided at this federally qualified health center include adult medical care, dental, mental health, addiction treatment, social work, nutrition, health education, and pharmacy services. Specialty hepatitis C treatment services are offered on-site.

For the purpose of the current study, all participants were recruited by either provider referral or via waiting room fliers, which included contact information for study personnel. Participants that contacted study personnel via telephone were given an appointment to conduct an in-person interview. Study personnel were available once weekly in the health center to conduct study interviews. Providers were made aware of the study during morning staff meetings and referred interested patients directly to the study personnel on the day of interviewing. Eligibility criteria included self-reported current tobacco use (verified with standardized tobacco use questions in the interview), available hepatitis C serology, and willingness to participate in the study interview. Once referred by provider or fliers, study personnel verified these characteristics. Individuals who met these criteria and provided informed consent completed an in-person, interviewer-administered questionnaire in a private room with a physician (AF, SM), psychology graduate student (HE, KSS), or pre-medical college student (AB-T). Responses were recorded in pencil and paper format by the interviewers. In the occasional instances when a participant did not understand a question, the interviewer explained the meanings of words, but avoided expressing opinions or judgments that could affect responses. Biochemical verification of smoking status was not performed. The interview lasted approximately 60 minutes and included questions about demographics, and general medical, psychiatric, and substance use histories. Hepatitis serologies were verified in the electronic medical record. All positive hepatitis C serologies were confirmed with a detectable hepatitis C viral load. Patients were considered hepatitis C-infected if they had a positive hepatitis C antibody assay and a detectable hepatitis C viral load, and patients were considered hepatitis C-uninfected if their most recent hepatitis C antibody assay was negative. The investigators based the interview content upon the recommendations for assessment to inform smoking cessation treatment in The Tobacco Dependence Treatment Handbook.¹⁶ This battery of scales is meant to guide the provider in developing an optimal management plan for his/her patients who smoke by measuring a number of key domains known to influence smoking and quitting behaviors based on Social Cognitive Theory,¹⁷ which views human behavior as the end result of the interactions between internal cognitive, affective, biological and behavioral factors, and external social/environmental influences. The various domains of tobacco use behaviors, attitudes, and psychosocial factors explored are summarized in Table 1. All participants received a $20 giftcard incentive for their participation.

Table 1.

Summary of Smoking Behavior Measures

Measure	Number of items	Rating schema	Explanation	Reference
Modified Fagerström Test for Nicotine Dependence	6	0–3 to questions pertaining to level of dependence on nicotine [Score = sum of responses]	A standard measure of nicotine dependence	Ref.18
Abrams and Biener Readiness to Quit Ladder	1	1 = “I have no interest in quitting” to 10 = “I have quit smoking”	A standard measure of readiness to quit	Ref.19
Reasons for Quitting	12	1 = “Not at all true” to 5 = “Extremely true” [Score = mean of responses]	Measures attitudes about intrinsic (eg, “To feel in control of my life”) and extrinsic (eg, “So my house or car won’t smell”) reasons for quitting	Ref.20
Self-Efficacy and Temptations	20	1 = “Not at all tempted” to 5 = “Extremely tempted” [Score = mean of responses]	Measures an individual’s self-estimation of his/her ability to resist smoking in various tempting situations including Positive Affect/ Social Situations (eg, “When I am happy and celebrating”), Negative Affect Situations (eg, “When I am very angry about something or someone”), and Habit/Craving Situations (“When I first get up in the morning”)	Ref.21
Locus of Control	25	1 = “Strongly disagree” to 6 = “Strongly agree” [Score = mean of responses]	Measures beliefs about where/with whom control of smoking behavior resides, Intrapersonal (eg, “I feel powerless to prevent myself from smoking when I am anxious or unhappy”), Interpersonal (eg, “Oftentimes other people drive one to smoke”), and Fate (eg, “Without the right breaks one cannot stop smoking”)	Ref.22
Smoker and Abstainer Self-Concept	9	1 = “Strongly disagree” to 10 = “Strongly agree” [Score = mean of responses]	Measures degree to which an individual identifies his/ herself as a smoker or an abstainer	Ref.23
Partner Interaction Questionnaire	20	0 = “Never” to 4 = “Very often” [Score = sum of responses]	Measures the individual’s prediction of frequency of positive and negative reinforcement for the attempted quit that his/her social supporter will offer	Ref.24
Perceived Risks and Benefits of Smoking	34	0 = “Not at all” to 2 = “A lot” [No aggregate scoring. Responses were compared by individual item]	A measure developed by the investigators to assess how much an individual believes smoking helps or harms different aspects of physical (including hepatitis C) or psychological health.	N/A
Importance of Smoking Cessation	1	1 = “Not important” to 10 = “Very important”	A measure assessing personal importance of quitting to a smoker	Ref.25
Interest in Tobacco Treatments	7	0 = “No” and 1 = “Yes” [No aggregate scoring. Responses were compared by individual item]	A measure developed by the investigators to assess interest in various tobacco treatment strategies	N/A
Advice from Primary Care Provider	5	0 = “No” and 1 = “Yes” [No aggregate scoring. Responses were compared by individual item]	A measure developed by the investigators to assess types of prior advice to quit smoking received from one’s primary care provider	N/A
Primary Care Post- Traumatic Stress Disorder Screen	4	0 = “No” and 1 = “Yes” [Score=sum of responses]	A primary care tool that assesses for the presence of post-traumatic stress disorder	Ref.26
Brief Symptom Inventory-18	18	0 = “Not at all” to 4 = “Extremely” [Score = mean of responses]	Measures symptom levels in the depression, anxiety, and somatization domains	Ref.27
Modified UCLA Loneliness Scale	10	1 = “Never” to 4 = “Often” [Score = sum of responses]	Measures frequency with which individuals experience symptoms of loneliness	Ref.28
Pain Inventory	1	0 = “No pain” to 10 = “The worst pain experienced ever”	Assesses highest level of pain experienced in the past week	Ref.29
Current Life Stressors	6	0 = “None” to 3 = “Major” [No aggregate scoring. Responses were compared by individual item]	A measure developed by the investigators to assess levels of current stress in various domains	N/A

Questionnaire responses were entered into a secure database according to the coding schema detailed in Table 1, and they were analyzed using SPSS 22.0 for Windows. For variables not listed in Table 1, continuous data were entered directly as numbers, categorical variables were assigned numeric response values and entered as numbers, dichotomous Yes/No variables were entered as 1/0, respectively, and text responses were entered as strings. For continuous variables, means ± SDs are presented. Comparisons of proportions were accomplished using chi-squared analysis or Fisher’s exact test. Comparisons of means were accomplished using Student’s t test or Mann-Whitney U test for ranked data or variables that violated normality and equivalence of variance assumptions. For all analyses, the independent variable was hepatitis C serostatus and the dependent variables were the characteristics listed in Tables 2 and 3. The lower and upper boundaries of 95% confidence intervals are presented for both ORs in comparisons of proportions and for mean differences in comparisons of means. An α ≤ 0.05 was considered statistically significant and all analyses were two-tailed.

Table 2.

Sociodemographic and Clinical Characteristics of the Study Sample

Characteristic	Hepatitis C+ (N = 40)	Hepatitis C− (N = 37)	OR (95% CI)	p
Age	54.9±8.2	48.0±13.5	6.9 (1.8–12.1)a	.01
Gender
Male	25 (62.5%)	19 (51.4%)	NA	.40
Female	15 (37.5%)	17 (45.9%)
Transgender	0 (0.0%)	1 (2.7%)
Latino ethnicity	20 (50.0%)	11 (29.7%)	2.36 (0.93–6.04)	.07
Black race	16 (40.0%)	22 (59.5%)	0.46 (0.18–1.13)	.08
Education
High school graduate or less	30 (75.0%)	21 (56.8%)	2.28 (0.87–6.02)	.09
Some college	10 (25.0%)	16 (43.2%)
Employment status
Employed	8 (20.0%)	6 (16.2%)	1.29 (0.40–4.15)	.67
Unemployed, retired, or disabled	32 (80.0%)	31 (83.8%)
Married or domestically partnered	10 (25.0%)	15 (40.5%)	0.49 (0.19–1.29)	.15
Financial status
Comfortable with some extras	7 (17.5%)	13 (35.1%)	NA	.29
Enough but no extras	16 (40.0%)	9 (24.3%)
Have to cut back	6 (15.0%)	5 (13.5%)
Can’t make ends meet	11 (27.5%)	10 (27.0%)
Mode of hepatitis C acquisition
Injection drug use	20 (50.0%)	NA	NA	NA
Sexual contact	11 (27.5%)
Transfusion	4 (10%)
Other/unknownb	5 (12.5%)
Comorbidities (past or present)
Depression	31 (77.5%)	21 (56.8%)	2.62 (0.98–7.04)	.05
Anxiety disorder	24 (60.0%)	19 (51.4%)	1.42 (0.58–3.51)	.45
Diabetes mellitus	6 (15.0%)	4 (10.8%)	1.46 (0.38–5.63)	.59
Hyperlipidemia	6 (15.0%)	8 (21.6%)	0.64 (0.20–2.06)	.45
Hypertension	17 (42.5%)	8 (21.6%)	2.68 (0.98–7.30)	.05
Asthma	14 (35.0%)	13 (35.1%)	0.99 (0.39–2.54)	.99
HIV infection	11 (27.5%)	7 (18.9%)	1.63 (0.55–4.77)	.37
Other substance use
Marijuana
Ever	35 (87.5%)	24 (64.9%)	3.79 (1.20–12.0)	.03
Current	15 (37.5%)	12 (32.4%)	1.25 (0.49–3.20)	.64
Cocaine
Ever	35 (87.5%)	18 (48.6%)	7.39 (2.37–23.0)	<.001
Current	10 (25.0%)	4 (10.8%)	2.75 (0.78–9.70)	.14
Heroin
Ever	29 (72.5%)	9 (24.3%)	8.20 (2.95–22.8)	<.001
Current	5 (12.5%)	1 (2.7%)	5.14 (0.57–46.3)	.20
Methadone
Ever	26 (65.0%)	5 (13.5%)	11.9 (3.78–37.3)	<.001
Current	13 (32.5%)	1 (2.7%)	17.3 (2.14–141)	.001
Alcohol
Ever	29 (72.5%)	29 (78.4%)	0.73 (0.26–2.07)	.55
Current	10 (25.0%)	22 (59.5%)	0.23 (0.09–0.60)	.002
Current mental health treatment	9 (22.5%)	3 (8.1%)	3.29 (0.82–13.3)	.12

CI = confidence interval; NA = not applicable; OR = odds ratio.

^aMean difference with 95% CIs.

^bIncludes three subjects who reported tattoos as the source of hepatitis C.

Table 3.

Behavioral Characteristics of the Study Sample

Characteristic	Hepatitis C+ (N = 40)	Hepatitis C− (N = 37)	OR or mean difference (95% CI)	p
Other tobacco/nicotine use
Cigar	5 (12.5%)	5 (13.5%)	0.91 (0.24–3.45)	.90
Pipe	0 (0.0%)	2 (5.4%)	NA	.23
Chewing tobacco	0 (0.0%)	1 (2.7%)	NA	.48
E-cigarette	3 (7.5%)	4 (10.8%)	0.67 (0.14–3.21)	.71
Daily cigarette consumption	8.8±5.8	10.6±10.0	1.8 (−2.0–5.5)	.85
Age at first cigarette	15.8±7.6	16.4±6.1	1.6 (−2.6–3.7)	.71
Lives with another smoker	10 (25.0%)	16 (43.2%)	0.44 (0.17–1.15)	.09
FTND scorea	4.33±2.06	4.03±3.04	0.30 (−0.90–1.49)	.68
Motivation to quit	6.15±2.09	6.46±1.74	0.31 (−0.57–1.19)	.49
Locus of control
Intrapersonal	3.65±1.30	2.80±1.19	0.85 (0.29–1.42)	.004
Interpersonal	3.91±1.26	3.62±1.17	0.29 (−0.26–0.84)	.30
Fate	3.46±1.26	3.27±1.20	0.19 (−0.37–0.75)	.50
Total	3.58±1.01	3.06±0.94	0.52 (0.07–0.96)	.02
Reasons for quitting
Intrinsic	3.74±0.95	4.04±0.83	0.30 (−0.11–0.71)	.15
Extrinsic	2.79±1.00	2.76±1.12	0.03 (−0.45–0.51)	.91
Self-concept
Smoker	4.30±2.74	3.41±2.63	0.89 (−0.34–2.11)	.15
Abstainer	7.23±2.46	7.62±2.46	0.40 (−0.72–1.52)	.48
Self-efficacy
Positive affect	3.35±1.05	3.14±1.13	0.21 (−0.29–0.70)	.41
Negative affect	4.00±1.01	3.81±1.09	0.19 (−0.29–0.67)	.43
Habit/craving	3.47±0.95	3.06±1.22	0.41 (−0.08–0.91)	.10
Total	3.68±0.83	3.44±0.89	0.24 (−0.15–0.63)	.23
Brief Symptom Inventory-18
Depression	1.15±1.12	0.73±1.06	0.43 (−0.07–0.92)	.09
Anxiety	1.09±1.08	0.91±1.17	0.18 (−0.33–0.69)	.49
Somatization	1.06±0.83	0.79±1.04	0.27 (−0.15–0.70)	.21
PTSD Screen Positive	7 (17.5%)	10 (27.0%)	0.57 (0.19–1.71)	.31
Pain during the past week	4.93±3.97	4.87±4.11	0.06 (−1.77–1.90)	.95
Partner Interaction Questionnaire
Positive behaviors	30.70±9.29	32.44±9.04	1.74 (−2.54–6.02)	.42
Negative behaviors	22.11±11.34	26.08±9.78	2.48 (−0.98–8.92)	.11
Perceived benefits of smoking
Controls pain	12 (30.0%)	5 (13.5%)	2.74 (0.86–8.75)	.10
Fights the hepatitis C virus	9 (22.5%)	0 (0.0%)	NA	.002
Helps fight infections	5 (12.5%)	1 (2.7%)	5.14 (0.57–46.3)	.20
Compared to patients like yourself, likelihood that smoking will lead to your death…
More likely	9 (22.5%)	17 (45.9%)	0.34 (0.13–0.91)	.03
Same or less likely	31 (77.5%)	20 (54.1%)
Personal importance of quitting smoking	8.30±2.83	8.75±2.88	0.45 (−0.86–1.76)	.50
Loneliness score	22.6±8.08	20.5±9.37	2.03 (−1.93–5.99)	.31

CI = confidence interval; NA = not applicable; OR = odds ratio; PTSD = post-traumatic stress disorder.

^aModified Fagerström Test for Nicotine Dependence.

Informed consent was obtained in writing from each participant. The study conformed to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the Albert Einstein College of Medicine Institutional Review Board.

Results

Seventy-seven participants completed the interview between June 2014 and January 2016. Forty were hepatitis C seropositive and 37 were hepatitis C seronegative.

Sociodemographic Profile, Clinical History, and Other Substance Use

The study population was middle-aged and overwhelmingly ethnic/racial minority (Table 2). Hepatitis C-infected participants were significantly older than hepatitis C-uninfected participants (54.9 vs. 48.0) and reported lower educational attainment that did not attain statistical significance in our study sample. Hepatitis C-infected were more likely than hepatitis C-uninfected participants to report histories of depression and hypertension.

History of other substance use was near universal in hepatitis C-infected smokers, and these rates significantly exceeded those of hepatitis C-uninfected participants for marijuana, cocaine, heroin, and methadone. Hepatitis C-infected smokers reported higher rates of current use for all non-alcoholic substances queried, but these differences did not achieve statistical significance except for methadone. Hepatitis C-uninfected individuals reported significantly higher rates of current alcohol use.

Table 3 details the smoking-related behavioral characteristics of the study sample as summarized in the following sections.

Tobacco Use History and Behaviors

Seventy-six of the 77 participants were current cigarette smokers (one participant was a current cigar smoker who did not use cigarettes). Among PLHC smokers, 12.5% also used cigars, 7.5% used e-cigarettes, and none used a pipe or chewing tobacco. There was no difference in other tobacco or e-cigarette usage patterns according to hepatitis C serostatus. PLHC smokers consumed an average of 8.8±5.8 cigarettes per day and this number did not differ significantly from hepatitis C-uninfected smokers. Hepatitis C-infected smokers were almost twice as likely to reside with another smoker than hepatitis C-uninfected smokers (42.3% vs. 25%), although this difference did not attain statistical significance. Utilization of prior quit assistance by PLHC smokers was reported as follows: 70% reported trying to quit “cold turkey,” 52.5% with nicotine replacement therapy, 15% with oral pharmacotherapy (bupropion or varenicline), 15% with acupuncture, 12.5% with group therapy, 7.5% with the telephone quitline, and 5% with individual counseling. None reported using an online tobacco treatment program. Utilization of prior quit assistance did not differ according to hepatitis C serostatus.

Nicotine Dependence and Readiness to Quit

There were no significant differences in nicotine dependence between groups according to hepatitis C serostatus. Among those with hepatitis C, 11 (27.5%) scored in the highly nicotine dependent range. There were also no differences in readiness to quit scores. Twenty-one (52.5%) of hepatitis C-infected smokers expressed their intention to quit within the next 30 days.

Intrinsic and Extrinsic Motivation to Quit

There were no significant differences in the motivation to quit subscales according to hepatitis C serostatuts.

Self-efficacy and Temptations

Hepatitis C-infected smokers reported lower self-efficacy (higher score indicates lower self-efficacy) to resist smoking across all domains, although none of these differences achieved statistical significance. The largest difference observed between groups was in the habit/craving subscale (eg, “How tempted to smoke will you be when you first get up in the morning?”).

Locus of Control

Hepatitis C-infected smokers exhibited significantly higher mean scores on the intrapersonal locus of control subscale, but there were no significant differences in the interpersonal or fate locus of control subscales according to hepatitis C serostatus.

Smoker and Abstainer Self-concept

There were no significant differences in smoker or abstainer self-concept according to hepatitis C serostatus.

Social Support

There were no significant differences in anticipated social supporter positive or negative behaviors according to hepatitis C serostatus.

Perceived Risks and Benefits of Smoking

More PLHC smokers reported pain control as a benefit of smoking (30.0% vs. 13.5%), but this difference did not achieve statistical significance. Almost a quarter (22.5%) of the hepatitis C-infected smokers thought that smoking helped to fight the HCV (none of the hepatitis C-uninfected group endorsed this belief). When asked to estimate the risk that smoking would cause his/her own death as compared to other patients in the clinic similar to him/herself (“More likely,” “About the same chance,” or “Less likely”) hepatitis C-infected smokers were more likely to minimize their comparative risk than hepatitis C-uninfected smokers, OR = 2.92 (1.09–7.81).

Personal Importance of Smoking Cessation

The self-rated importance of smoking cessation was high, exceeding eight on a scale of one to ten, and there were no significant differences in this measure according to hepatitis C serostatus.

Interest in Smoking Cessation Interventions

When asked which cessation strategies they would be interested in using, PLHC smokers reported interest in the following non-mutually exclusive categories: nicotine replacement therapy 70%, individual counseling 55%, group counseling 42.5%, oral pharmacotherapy 37.5%, quitline 35%, and online tobacco treatment 17.5%. There were no significant differences in interest in smoking cessation strategies according to hepatitis C serostatus.

Advice From Primary Care Providers About Smoking Cessation

Ninety percent of PLHC smokers stated that their provider had spoken with them about quitting, 77.5% had been given a quit-smoking brochure, 72.5% reported receiving recommendations for pharmacotherapy, 40% were advised to call the quitline, and 17.5% were referred to a smoking cessation program. There were no significant differences in receipt of advice from primary care providers according to hepatitis C serostatus.

Anxiety and Depression

There were no significant differences in anxiety and somatization symptomatology according to hepatitis C serostatus, and the higher depression scores exhibited by hepatitis C-infected smokers failed to attain statistical significance (p = .09).

Loneliness

Six hepatitis C-infected participants (15.0%) and six hepatitis C-uninfected participants (16.2%) scored in the extremely lonely range, that is, score > 30,²⁸ but there was no significant difference in overall loneliness score according to hepatitis C serostatus.

Pain

There was no significant difference in worst pain experienced in the past week according to hepatitis C serostatus.

Current Life Stressors

Among PLHC smokers, 80% reported at least some financial stress, including 37.5% who reported major financial stress; 80% reported at least some health-related stress, including 30% who reported major health-related stress; 62.5% reported at least some relationship-associated stress, including 27.5% who reported major relationship-associated stress; 54.5% reported at least some personal safety-related stress, including 14.3% who reported major personal safety-related stress; 50% reported at least some job-related stress, including 10% who reported major job-related stress; and 13.0% reported at least some sexual abuse-related stress, including 2.6% who reported major sexual abuse-related stress. There were no significant differences in current life stressors according to hepatitis C serostatus.

Discussion

The present study analyzed the responses from detailed questionnaires completed by a group of hepatitis C-infected and uninfected smokers receiving care at a clinical site in the Bronx, New York. The smokers living with hepatitis C in our patient sample were middle-aged, ethnic/racial minority, with low educational attainment, high rates of unemployment, and heavy burdens of substance use and psychiatric disease. They exhibited high levels of motivation to quit, identified the locus of control of smoking behaviors within themselves, but also attributed a variety of positive effects to their tobacco use, some of which reflected mistaken belief systems (eg, “Smoking helps fight the HCV”).

The prevalence of chronic hepatitis C in the US population is a dynamic statistic and the availability of highly effective, curative therapy will have a profound impact on the epidemiology of this infection in the years to come.³⁰ The prevalence of active tobacco use among PLHC in the United States is not precisely defined. The few population-based data available, however, suggest that well over half of PLHC smoke cigarettes,^3,4 and this statistic is substantiated by additional, more geographically restricted studies.^31,32 A conservative estimate of the PLHC smoker population in the United States is 1–1.5 million individuals, representing 2.5%–3.8% of all US smokers.³³

Psychiatric and substance use comorbidities are extremely common in PLHC.⁵ Our findings suggest that these comorbidities are indeed hyperprevalent in the subset of PLHC who smoke. This reality has significant implications for the tobacco treatment community. Psychiatric illness and substance use are both factors that define “complicated” smokers who are unlikely to respond to brief interventions and should be referred for more intensive tobacco treatment.⁶ Unfortunately, little is known about the psychobehavioral mechanisms (triggers/causes) of tobacco use in PLHC, so the ability of tobacco treatment practitioners to appropriately target their message may be limited. The data presented herein may prove helpful in optimizing tobacco treatment for this highly vulnerable population.

Past or present depression was identified at higher rates in PLHC smokers and this group also exhibited higher depression scores on the Brief Symptom Inventory scale. Tobacco treatment, both pharmaco- and behavioral therapy, may be administered safely and effectively to depressed individuals, but practitioners and patients must be attuned to any acute worsening in psychiatric status during the quitting process.⁷

Tobacco treatment may also be delivered effectively to substance using populations.⁷ The majority of our subjects were ex-substance users. Their prior successes in overcoming addictions to other substances could be highlighted as sources of encouragement in a targeted intervention. Coping skills that were used to successfully stop and remain abstinent from other addictions can be identified and used to develop coping strategies to help quit tobacco. Unfounded fears over the possibility of relapses during the quitting process could also be assuaged with the presentation of evidence to the contrary.⁷

The sociodemographic profile of our sample indicates that tobacco treatments for PLHC smokers should speak to a middle-aged, ethnic minority audience. Since most smokers begin using cigarettes as teenagers, our mean age of 54.9 translates into approximately four decades of smoking experience on average. It is important for such smokers to understand that cessation is possible even after many years of tobacco use, and testimonials may be employed to attest to this. Educational levels were lower in hepatitis C-infected smokers, so curricula and written materials for distribution should be sensitive to the possibility of health literacy challenges.

There were a few notable differences between hepatitis C-infected and uninfected smokers in the behavioral domains and belief systems that we explored. PLHC smokers scored significantly higher on the intrapersonal locus of control subscale. This is indicative of a higher level of belief that internal factors rather than external influences or fate are the leading determinants of smoking behaviors. Prior research shows that higher intrapersonal locus of control scores predict successful cessation.³⁴ It is possible that past experiences overcoming other substance addictions may counter the “once a smoker always a smoker” mentality associated with greater externalization in the locus of control measure.³⁵ It is also possible that enthusiasm surrounding new treatments for hepatitis C, which place PLHC’s hepatitis infection status in their own hands, may reinforce the view that PLHC are in control of their own health-related destinies. Tobacco treatment messages targeting PLHC should aim to reinforce the belief that the ultimate agent of cigarette smoking behaviors is the smoker him or herself.

Although the findings did not achieve statistical significance in this small sample, PLHC smokers scored lower on total abstinence self-efficacy and on all of the subscales, especially in the habit/craving domain. A major goal of many counseling interventions is the fortification of self-efficacy, that is, one’s estimation of his/her own ability to resist the urge to smoke in different situations, and these results may suggest that tobacco treatment for PLHC should emphasize strategies to avoid habit/craving situations or to respond to them without lighting up a cigarette. Almost a third of PLHC smokers viewed positively the analgesic effects of tobacco use compared to 13.5% of hepatitis C-uninfected smokers. Interventions should emphasize the availability of healthier and more effective methods of controlling pain symptoms.

It is remarkable that almost a quarter of our hepatitis C-infected sample expressed the belief that cigarette smoking helps in “fighting the HCV.” This finding is strikingly similar to that reported by us in HIV-infected smokers with regard to their HIV virus.³⁶ Our hepatitis C-infected participants also tended to minimize the likelihood that smoking would lead to their demise in comparison to our hepatitis C-uninfected subjects. Assessment of comparative risk, that is, how likely an individual thinks s/he is to die from a given behavior compared to other individuals similar to him or herself, is a distinct and important dimension of risk perception.³⁷ Smokers who understand that they are engaging in a risky behavior may downplay their overall risk through comparative risk perception,³⁷ and our findings suggest that PLHC smokers do so. Although our PLHC smokers professed relatively high motivation to quit, it is clear that effective tobacco treatment strategies for them must include accurate information to dispel misconceptions about imagined benefits of smoking and about the absolute health risks of their tobacco use.

There are significant parallels between the smoking epidemics embedded within both the hepatitis C and the HIV-infected communities in the United States. Both disproportionately involve poor, ethnic minority subsets of the population, both suffer with pervasive psychiatric and other substance use comorbidities, and both have entered eras of effective antiviral therapy, although the advent of highly active therapy for HIV antedated that for hepatitis C by more than a decade. Just as cigarette smoking has emerged as the leading killer among persons living with HIV now that the viral infection may be effectively controlled,³⁸ it is reasonable to predict that the broader application of hepatitis C treatments will unmask the harm of chronic tobacco use in PLHC, and it will similarly emerge as a leading cause of morbidity and mortality. Some experts predict the end of the hepatitis C epidemic in the United States in the 2030’s,³⁰ but without drastic action the tobacco use epidemic is destined to live on in this vulnerable group even after the virus is eradicated. While our findings indicate that PLHC-smokers are interested in quitting, we are not aware of any tobacco treatment programs that are specifically designed to help actualize their intentions. Their care providers do not ignore the issue, but many treatment modalities, such as counseling, quitlines, and online programs, are severely underutilized. There is a great need for more to be done in both quantity and quality. Research such as this may prove helpful in optimizing tobacco treatment strategies for PLHC smokers in the future.

Certain limitations of the current study require mention. The sample size was small, and it is possible, indeed likely, that a larger sample may have revealed additional biobehavioral associations with smoking in PLHC. With a total population of 77 participants, the potential for beta-errors in our analyses was substantial. The sample was derived from a single care site in the South Bronx, and may not be generalizable to the broader US PLHC population. Moreover, the fact that the participants were recruited from a clinic population selected for a group of individuals who were in care and may not be representative of out-of-care smokers. The face-to-face data collection strategy, while minimizing missing data points and mitigating the challenges of misunderstood questions and lapses in participant attention, is subject to social desirability bias, particularly in the realm of substance use research. Although smoking behavior self-report is generally reliable, especially when current tobacco use is acknowledged,³⁹ we did not verify smoking status biochemically. We also did not repeat negative hepatitis serologies at the time of study entry. These limitations do introduce a small risk of misclassification bias into the study.

Tobacco use in persons living with hepatitis C is a significant public health challenge, and little has been done to address it. We have entered an era that will see large numbers of PLHC interacting intensively with the health care system in order to access antiviral therapy. This represents a golden, and perhaps singular, opportunity to intervene with these patients in other areas of health promotion. It makes little sense from the public or personal health standpoint to spend $1000 per pill on hepatitis C treatment yet ignore the lethal tobacco use epidemic embedded in this population. The study that we present herein represents a first step toward a better understanding of the drivers of tobacco use in PLHC. Our findings may be useful in guiding tobacco treatment strategies for PLHC smokers. Larger observational and treatment studies will be necessary to better define the problem and to optimize future therapeutic options.

Funding

JS’s effort on this project was supported in part by awards 1R01DA036445, 1R01CA192954, and 1R34DA037042 from the National Institutes of Health. BN’s effort of this project was supported in part by award 1K23DA039060 from the National Institutes of Health.

Declaration of Interests

AHL has served on the advisory boards of Gilead Sciences, Merck Pharmaceuticals, BMS Pharmaceuticals, and Janssen Pharmaceuticals. None of the other authors have potential conflicts of interest to report.