Smoking Habits in Human Immunodeficiency Virus-Infected People from Italy: A Cross-Sectional Analysis of the STOPSHIV Cohort

Abstract

In a nationwide Italian sample of people living with HIV (PLWH), the prevalence of smoking, nicotine dependence, propensity to stop smoking, and cardiovascular profile were investigated. The nicotine dependence by Fagerström test and the propensity to stop according to the stages of change were evaluated. Associations between smoking habits and patients' characteristics were analyzed using an unconditional logistic regression model. Among 1,087 PLWH (age 47.9 ± 10.8 years, men 73.5%), the prevalence of current smokers was 51.6%. The median of Fagerström test was 4 (interquartile range 2–6); 60.1% of the smokers were in precontemplation, 17.6% in contemplation, 18.7% in preparation, and 3.6% in action. In a logistic multivariate model, current smoking was associated with male sex, being divorced/widowed, Caucasian ethnicity, dyslipidemia, atherosclerotic cardiovascular disease risk, psychiatric comorbidity, hepatitis C virus infection, and alcohol abuse. Low high-density lipoprotein cholesterol level was associated with high nicotine dependence. More than 50% of PLWH were current smokers, one-third of them showed a high or very high degree of dependence. Our findings draw attention to the need of smoking cessation strategies for PLWH.

Introduction

Tobacco use is a leading cause of preventable illness and death for all individuals, but it is even more of a concern for people living with HIV (PLWH), who tend to smoke more than the general population. Well treated PLWH may lose more life years through smoking than through HIV.¹ Cigarette smoking increases morbidity and mortality for both HIV and non-HIV-related diseases. Lung and cardiovascular diseases (CVDs) are major concerns, as tobacco use and HIV together may accelerate the development of chronic obstructive pulmonary disease (COPD), lung cancer, and death,^2,3 and smoking is associated with a higher risk of myocardial infarction in the HIV-infected patients than in the general population as well.⁴ Characterizing and appropriately managing smoking habits are, therefore, a clinically relevant issue in PLWH.

Current clinical guidelines mandate preventive interventions for lifestyle modification, and are designed to improve the prognosis in HIV-infected patients.⁵ Little is known about the characteristics of Italian PLWH smokers, and it is unclear how many of them are really interested in quitting. In other countries, smoking has been associated with age, educational level, illicit drug use, and heavy drinking among PLWH.^6
–8 However, up to date no study about cigarette smoking, nicotine dependence, and related factors has been conducted among PLWH in Italy.

In this study, we aimed at investigating prevalence of smokers, propensity to stop smoking according to the stages of change, nicotine dependence, and quitting rate in HIV-infected patients referring to a network of 10 Infectious Diseases Clinics in Italy. As a secondary outcome, we analyzed factors associated with smoking status and cardiovascular (CV) profile.

Materials and Methods

The study was conducted by the Coordinamento Italiano per lo Studio di Allergie e Infezione da HIV (CISAI, Italian coordinating group for the study of allergies and HIV infection). Patients were enrolled in the “PROJECT STOP Smoking in HIV patients” (see Appendix). A total of 1,087 adult PLWH attending scheduled or unscheduled outpatient visits at one of the centers of the CISAI group were enrolled in the study from July 2014 through September 2016. This article is a cross-sectional analysis of baseline data from the cohort study.

Smoking habits, nicotine dependence, and propensity to stop according to the stages of change were investigated by a standardized questionnaire including the 6-item Fagerström test for Nicotine Dependence (FTND) used to assess the level of nicotine dependence.⁹ A score of 0–2 corresponds to very low nicotine dependence, 3–4 to low dependence, 5 to moderate dependence, 6–7 to high dependence, and 8–10 to very high dependence.¹⁰ A score of ≥5 on the 6-item FTND was defined as moderate–very high dependence. If an item was missing, the lowest score was assigned by default, so it was likely that in some patients the total score was underestimated.

Other data regarding PLWH, such as anthropometric measures, blood pressure (BP), history of diabetes, chronic hepatitis, HIV stage according to the Centers for Disease Control (CDC) classification, use of antihypertensive drugs, and lifestyle counseling were recorded using a standard data collection form with the same methods reported in another study by the CISAI study group.¹¹ Clinical symptoms of COPD such as dyspnea, chronic cough, or sputum production were also collected.¹² Information from laboratory tests included absolute CD4+ T-lymphocyte count, fasting total cholesterol (TC), low-density lipoprotein cholesterol, and high-density lipoprotein cholesterol (HDL-C), triglycerides, plasma glucose concentration, and creatinine levels. Scheduled parameters were collected by a physician.

Among other CV risk factors, hypertension was defined as office systolic BP of ≥140 mmHg and/or a diastolic BP of ≥90 mmHg, or receiving antihypertensive therapy at the time of the index examination.¹³ Office BP was measured by a physician in the outpatient clinic and measurement procedures were standardized before starting enrollment as previously reported.¹¹ The diagnosis of diabetes was based on standard international criteria.¹⁴ Dyslipidemia was defined as TC >200 mg/dL or triglycerides >150 mg/dL or use of statins or other lipid-lowering drugs. Psychiatric comorbidity was defined as having a psychiatric diagnosis and/or being on chronic pharmacological therapy with drugs for anxiety, depression, or psychosis.

Alcohol abuse was defined as alcohol drinking >35 g for day (i.e., 3 U or more, calculated as 1 U = 125 mL wine or 330 mL beer or 30 mL spirits, all containing ∼12 g of ethanol.

The 10-year risk for CVD was estimated using the Pooled Cohort Equations atherosclerotic cardiovascular disease (ASCVD) risk.¹⁵ According to Mdodo et al.,⁷ “never smokers” were defined as persons who reported smoking <100 cigarettes in their entire lifetime, “current smokers” were defined as persons who reported smoking 100 cigarettes or more during their lifetime and currently smoked every day or some days, and “former smokers” were defined as those who reported smoking at least 100 cigarettes during their lifetime, but currently were no longer smoking. The quit ratio, a measure of successful smoking cessation at the population level, was defined as the ratio of former smokers to ever smokers (i.e., the sum of current smokers and former smokers). The smoking history was evaluated as lifetime smoking exposure, it was quantified in “pack years,” where one “pack year” is defined as 20 cigarettes smoked per day for 1 year.¹⁶ Patients received regular follow-up visits monitoring HIV infection at least every 6 months at the outpatient Infectious Diseases clinics.

All subjects provided informed consent to participate in the study, which was approved by the institutional ethics committee of the coordinating center (Ethics Committee of the Umbria Region) and participating centers. Confidentiality was assured because all patients' information was anonymized, and secured storage was prepared for both paper questionnaires and electronic data set.

Statistical analysis

Categorical and discrete variables were described as frequency (N) and percentage (%), continuous variables as mean and standard deviation (SD, mean ± SD) if normally distributed, and median and interquartile range (IQR) if not normally distributed. In the crude analysis, we used the Pearson or Mantel–Haenszel χ ² test (as appropriate) to assess the association between categorical variables. Means were compared using the analysis of variance and medians by means of Mann–Withney U test. Associations between smoking habits and patients' characteristics were evaluated using an unconditional logistic regression: odds ratio (OR) and 95% confidence interval (CI) for current and former smoking were calculated using never smokers as reference category. Furthermore, we also calculated ORs of former smokers as compared with current smokers. Continuous normally distributed variables were compared using a general linear model; if not normally distributed, square-root transformation was completed before including them in the general linear model. Multivariable analyses were performed to account for confounding factors. Potential confounders included in the multivariable analysis were age, sex, body mass index (BMI), family history of CVD, risk factor for HIV acquisition [intravenous drug use (IVDU), sexual route, and others], marital status, and education. To avoid the risk of over adjusting, ethnicity was excluded from the adjusted model because it was strongly associated with sex, education, BMI, and risk factor for HIV acquisition. Association of each infection variable and comorbidity with smoking status was estimated in this multivariable model, including them in the model equation as appropriate (i.e., correlated variables were included in turn).

Data analysis was conducted using the statistical software SAS/STAT for Windows (version 9.4).

Results

Overall, 1,087 HIV patients were enrolled from July 2014 to December 2016 in 10 participating centers. Mean age was 47.9 ± 10.8 years, higher in men (48.6 ± 11.1) than in women (46.0 ± 9.6). Men were 73.5% of the whole sample, and most patients were Caucasians (89.2%). The main risk factor for HIV acquisition was sexual route (72.1%), followed by IVDU (22.2%). Forty-three patients (4.0%) were antiretrovirals naive; in experienced subjects, median time of antiretroviral treatment was 9.3 (IQR 3.9–16.0) years.

Prevalence of current smoking was 51.6% and median of estimated pack-years was 21.8 (IQR 12.6–36.0). Pack-years >30 were reported by 35.6% of current smokers. The Fagerström test median value of was 4 (IQR 2–6). The propensity to stop smoking according to the stages of change is depicted in Figure 1: >60% of current smokers were in the precontemplation stage.

FIG. 1.

Propensity to stop smoking habit in current smokers, according to the stages of change.

Patients' main characteristics according to smoking habits are reported in Table 1. Groups of smoking habits varied in terms of age, sex, marital status, and education, ethnicity, BMI, previous CVD events, diagnosed diabetes and hypertension, ASCVD, psychiatric comorbidity, alcohol abuse (current and former), and hepatitis C virus (HCV) coinfection. With regard to HIV infection characteristics, smoking groups also showed significant risk factor differences for HIV acquisition, CDC stage, and overall antiretroviral therapy (ART) duration. The blood lipids profile was significantly less favorable in current than in never or former smokers.

Table 1.

Characteristics of Sample Population Enrolled in STOPSHIV (n = 1,087)

	Cigarette smoking
	Never		Current		Former
	N = 317	29.2%	N = 561	51.6%	N = 209	19.2%	p
Age, years (mean ± SD)	46.0 ± 11.6		47.0 ± 9.8		53.2 ± 10.3		<.0001
Sex							<.0001
Female	112	35.3	137	24.4	39	18.7
Male	205	64.7	424	75.6	170	81.3
Education							.02
College degree	67	21.1	147	26.2	50	23.9
High school diploma	139	43.8	274	48.8	106	50.7
Secondary school	108	34.1	134	23.9	52	24.9
Marital status							.04
Single	148	46.7	246	43.9	72	34.4
Stable relationship	128	40.4	203	36.2	95	45.5
Divorced	23	7.3	65	11.6	26	12.4
Widowed	7	2.2	16	2.9	7	3.3
Ethnicity							<.0001
Caucasian	242	76.3	528	94.1	200	95.7
Other	75	23.7	33	5.9	9	4.3
Unemployed	67	21.1	137	24.4	34	16.3	.048
BMI, kg/m² (mean ± SD)	25.1 ± 4.1		24.1 ± 4.1		25.4 ± 4.1		<.0001
Risk factor for HIV acquisition							<.0001
MSM	91	28.7	157	28.0	48	23.0
IVDU	15	4.7	179	31.9	47	22.5
Heterosexual	189	59.6	201	35.8	96	45.9
Transfused	3	0.9	2	0.4	2	1.0
Other	17	5.4	22	3.9	16	7.7
CDC stage							.79^a
A	151	47.6	241	43.0	104	49.8
B	84	26.5	171	30.5	37	17.7	.005^b
C	79	24.9	139	24.8	60	28.7	.84^b
Missing	3	0.9	10	1.8	8	3.8
Detectable HIV RNA (>40 cp/mL)	39	12.3	69	12.3	19	9.1	.38
Current CD4, cell/mm³							.41^a
<200	20	6.3	34	6.1	11	5.3
200–350	37	11.7	52	9.3	21	10.0	.81^c
>350	257	81.1	474	84.5	176	84.2	.85^c
Nadir CD4 < 200 cell/mm³	152	47.9	246	43.9	107	51.2	.46
ART treatment duration, years (median, IQR)	7.0	3.0–12.5	10.3	4.0–17.0	10.2	5.0–16.0	<.0001
Diabetes	13	4.1	32	5.7	19	9.1	.049
Treated hypertension	53	16.7	89	15.9	57	27.3	.0004
Dyslipidemia	156	49.2	313	55.8	130	62.2	.01
ASCVD (median, IQR)	2.5	1.0–6.0	7	4.3–11.3	5.3	2.4–9.3	<.0001
Psychiatric comorbidity	16	5.0	79	14.1	17	8.1	<.0001
Hepatitis C virus positive	29	9.1	191	34.0	56	26.8	<.0001
Alcohol abuse							<.0001
Current	3	0.9	41	7.3	10	4.8	.0001
Former	2	0.6	23	4.1	9	4.3	.006
COPD clinical symptoms (n = 975^d)
Cough	5	1.6	233	41.5	13	6.2	<.0001
Sputum production	3	0.9	199	35.5	19	9.1	<.0001
Dyspnea	12	3.8	119	21.2	27	12.9	<.0001
Total cholesterol, mg/dL	188 ± 39		185 ± 42		187 ± 43		.58
HDL-C, mg/dL	50 ± 15		46 ± 14		50 ± 15		.0003
LDL-C, mg/dL	113 ± 32		109 ± 34		110 ± 36		.29
Triglycerides, mg/dL	105 (77–154)		124 (87–184)		123 (89–175)		.0001
HDL-C/TC	0.27 ± 0.08		0.26 ± 0.08		0.27 ± 0.09		.03

All data are reported as number and percentage if not otherwise specified.

Mantel–Haenszel test.

Reference: stage A.

Reference <200 cells/mL.

One center did not record this information.

SD, standard deviation; BMI, body mass index; MSM, men having sex with men; IVDU, intravenous drug use; ART, antiretroviral therapy; ASCVD, atherosclerotic cardiovascular disease risk score (10 years); hepatitis C virus positive, antibodies positive for HCV; COPD, chronic obstructive pulmonary disease; CDC, Centers for Disease Control; ART, antiretroviral therapy; TC, total cholesterol, LDL-C, low-density lipoprotein-cholesterol; HDL-C, high-density lipoprotein-cholesterol; IQR, interquartile range.

In a multivariable regression model taking into consideration the covariates of age, sex, BMI, risk factor for HIV acquisition, and family history of CVD, we found that HDL-C and HDL-C/TC ratio were significantly lower in current than in never smokers (p = .044 and p = .040, respectively). Similarly, triglyceride levels were higher (p = .0003). When including years of ART in the equation, these differences were still significant. In this multivariable analysis, antiretroviral duration was not related with smoking status. Similarly, in a logistic multivariate model, with never smokers as the reference category (Table 2), current smoking remained positively associated with male sex, being divorced or widowed, Caucasian ethnicity, family history of CVD, dyslipidemia, ASCVD risk score, psychiatric comorbidity, HCV antibodies, and current and former alcohol abuse.

Table 2.

Multivariable Logistic Analysis: Associations Between Smoking Habits and Patients' Characteristics

	Current vs. never smokers		Former vs. never smokers		Former vs. current smokers
	aOR	95% CI	aOR	95% CI	aOR	95% CI
Age (by 5 years)	1.00	0.93–1.08	1.36	1.28–1.50	1.36	1.24–1.50
Sex female (ref. male)	0.70	0.49–0.99	0.51	0.31–0.85	0.90	0.57–1.92
Marital status (ref. stable relationship)
Single	1.26	0.89–1.80	0.93	0.59–1.47	0.72	0.48–1.08
Divorced/widowed	2.03	1.19–3.47	2.23	1.16–4.28	0.86	0.51–1.43
Unemployed (ref. N)	1.16	0.78–1.71	1.16	0.67–2.01	0.79	0.50–1.25
Ethnicity (ref. Caucasians)	0.28	0.17–0.47	0.30	0.14–0.64	1.13	0.49–2.57
BMI >25.0 (ref. <25.0)	0.69	0.50–0.96	1.11	0.74–1.68	1.52	1.06–2.19
Risk factor for HIV acquisition (ref. sexual)
IVDU	8.93	5.04–15.82	4.49	2.30–8.74	0.50	0.32–0.75
Other	0.81	0.38–1.74	1.22	0.52–2.85	1.85	0.82–4.21
CDC stage (ref. A)
B	0.99	0.88–1.46	0.41	0.23–0.71	0.46	0.29–0.75
C	0.91	0.61–1.37	0.72	0.44–1.19	0.79	0.51–1.23
Dyslipidemia (ref. N)	1.44	1.04–2.01	1.36	0.88–2.11	0.83	0.56–1.21
ASCVD^a (by one point)	1.13	1.08–1.17	1.07	1.03–1.11	0.97	0.94–0.99
Psychiatric comorbidity (ref. N)	2.89	1.55–5.39	1.66	0.71–3.90	0.50	0.27–0.92
Hepatitis C virus positive (ref. N)	2.79	1.68–4.63	2.17	1.12–4.21	0.82	0.52–1.30
Alcohol abuse (ref. N)
Current	5.77	1.69–19.74	4.76	1.12–20.31	0.83	0.39–1.77
Former	6.01	1.31–27.55	6.75	1.14–39.70	1.27	0.54–3.01
COPD clinical symptoms (n = 975^b)
Cough (ref. N)	25.93	10.32–65.19	1.56	0.46–5.27	0.07	0.04–0.13
Sputum production (ref. N)	31.11	9.68–100.1	4.05	1.08–15.10	0.14	0.08–0.24
Dyspnea (ref. N)	2.77	1.40–5.47	1.32	0.58–3.04	0.46	0.27–0.78

The values in bold represents p < .05.

Not age adjusted and sex adjusted.

One center did not record this information.

aOR, adjusted odds ratio; CI, confidence interval; ref, reference category; N, no.; ASCVD, atherosclerotic cardiovascular disease risk score (10 years); hepatitis C virus positive, antibodies positive for HCV.

As expected, symptoms of COPD were positively associated with current smoking. Comparing former and never smokers (Table 2), we observed a largely similar pattern of associations, although some variables lost statistical significance. The estimates regarding COPD symptoms were remarkably lower.

In a further analysis using current smokers as the reference category, being a former smoker was positively associated with older age and higher BMI; an inverse association was observed with IVDU, CDC stage B, and ASCVD score (Table 2). In addition, quitting was inversely associated with psychiatric comorbidities. All symptoms of COPD were significantly less frequent in former than in current smokers.

Overall, the quit ratio for smoking was 27.1% (95% CI 24.1–30.4%). Among current smokers, the dependence degree, depicted in Figure 2, was similar in both sexes. Overall, 176 (31.4%) subjects had high or very high dependence. When comparing smokers with high to those with low-moderate dependence, we found that the former were older (48.4 vs. 46.3 years, p = .02), less educated (34.9% vs. 19.2% with junior secondary, 20.0% vs. 29.5% with college degree, Mantel–Haenszel χ ² p = .0001), had more frequently IVDU as a risk factor for HIV acquisition (43.2% vs. 26.8%, p = .0005), and were in CDC stage B or stage C rather than stage A (37.2% and 29.6% vs. 29.2% and 23.2%, respectively, p = .003). Lastly, those with high dependence smoked 23.1 versus 12.5 cigarettes per day (p < .0001).

FIG. 2.

The dependence degree among current smokers by Fagerström test: very low (0–2 points), low (3–4 points), moderate (5 points), high (6–7 points), and very high (8–10 points).

When including these factors in a multivariable regression, we could confirm an independent association between high dependence and low education (adjusted OR, aOR, 3.14, 95% CI 1.79–5.51 for junior secondary education as compared with college degree) and IVDU (aOR 1.69, 95% CI 1.11–2.58 as compared with sexual route).

Finally, smokers with high/very high dependence showed a significantly lower level of HDL-C (46 vs. 49 mg/dL, p = .014), after considering age, sex, BMI, and risk factor for HIV acquisition as potential confounders.

Discussion

Our study contributes to the literature by providing, to our knowledge for the first time, detailed information on smoking habits among PLWH in Italy. The main results of this investigation demonstrated that in a sample of Italian HIV-infected patients seen in routine outpatient clinical care, with an average age ∼50 years and the majority receiving antiretroviral treatment, the overall prevalence of smoking habits was >50%; the level of tobacco smoking was markedly higher in HIV-infected patients than in the Italian general population,¹⁷ in which the prevalence was <20% in the same observational period.

These Italian data are in concordance with previously published research on PLWH from other countries. A recent meta-analysis¹⁸ reported a prevalence of current smoking at 36.3% (95% CI 28.0–45.4) among women and 50.3% (95% CI 44.4–56.2) among men. Consistently with previous research, in our study IVDU, history of major CV events, psychiatric comorbidity, and current or past alcohol abuse⁶ were all factors associated with cigarette smoking in HIV-infected patients. Thus, we can reasonably presume that factors related to HIV risk (such as illicit drug use, alcohol abuse, loneliness, depression, and anxiety) mediate the relationship between smoking and HIV infection.

Indeed, a subset of HIV-infected current smokers is exposed to additional major health risks by consuming alcohol and other substances, as is well documented in previous investigations.^19,20 In our study, the degree of nicotine dependence was high or very high (Fagerström test >5 points) in 31.4% of cases. This proportion was lower than that of a previous study from Austria and Germany by Brath et al.,²¹ which reported a Fagerström test >5 points in 48.4% of smokers in a sample of 221 PLWH. However, as previously stated, in some of our patients, this score might be underestimated.

Regarding HIV parameters, we did not find significant differences between smokers and nonsmokers, with respect to CDC stage, CD4 nadir, and undetectable HIV RNA. Other studies showed that cigarette smoking adversely affects the immunologic response to ART²² and has a negative association with antiretroviral adherence and positive association with viral load among people with HIV.²³ Conversely, in our sample, current smokers showed a low level of HDL-C and HDL-C/TC than never as well as former smokers. This finding is in line with recent reports showing that smokers have less favorable lipid profiles.^24,25 Smokers with high Fagerström test, that is, those with a more severe nicotine dependence, showed lower levels of HDL-C, a finding that is consistent with the recent investigation by Selya and Hesse.²⁵

Our results suggest that smoking may be even more detrimental to individuals with hyperlipidemia. Besides, mitochondrial dysfunction and chronic inflammation are induced through oxidative stress smoking.²⁶ Thus, all these factors negatively clustered for CVDs.

Moreover, in this study, the quit ratio for smoking was 27.1%. This was slightly worse than the quit ratio found in a U .S. study,⁷ reporting a quit ratio of 32.4% in HIV-infected patients and 51.7% in the general population. Unlike the report by Tesoriero et al.,¹⁹ in STOPSHIV most current smokers answered that they did not want to quit smoking. The majority of participants were in the “precontemplation” phase (60.1%), and only 18.7% and 3.6% were found to be in the “preparation” or in the “action” phase, respectively. Unfortunately, higher dependence levels are linked with lesser expected resolve to stop smoking.²⁷

Our findings suggest that there is an important unmet clinical need for programs to provide all HIV professionals with the knowledge base for navigating the increasingly complex demands of comprehensive HIV care. This is particularly true in light of the rising impact of HIV-associated non-AIDS conditions in the aging HIV population. Interventions are urgently needed, aiming at reducing codependence of both smoking and alcohol use, even if the expected success rate is low. Understanding and optimizing preventive care in HIV patients are essential in maintaining the substantial advances in prognosis for those subjects. Despite that HIV-infected patients are considered at high CV risk, they still tend to be undertreated for CV prevention; in this setting, counseling to quit smoking represents really a key strategy to prevent CV diseases.^28,29 In our analysis, the estimated risk for future CV events is significantly higher in current smokers than both never and former smokers, so PLWH need to be carefully monitored and counseled in a clinical setting regarding changing behavioral risk factors.

This study has implications for clinical practice. More than 35% of smokers among HIV-infected patients had >30 pack-years (46% of those aged >40 years), thus according to Italian guidelines they are theoretically candidates to annual chest tomography for lung cancer screening.³⁰

This study has a number of strengths. This study was deliberately inclusive, there was no selection, and a well-characterized multicenter sample of Italian HIV-infected outpatients, followed at Infectious Diseases clinics, were examined, most likely representative of the general assisted population at the same sites. We adopted a standardized procedure for all observed patients. We collected data on smoking habits and clinical information including HIV stage, ART, CD4, and HIV RNA.

However, several limitations need to be discussed. The cross-sectional design and the preliminary nature of our report that is the starting point of a longitudinal study are a primary limitation; the relatively low number of HIV participants older than 65 years on STOPSHIV study represent a second limitation; however, this reflects the current composition of this population.³¹ Additional limitation is the absence of laboratory markers of smoking status, our categorization was only based on patient's reported data; however, the smokers' prevalence levels correspond well with measured exhaled carbon monoxide levels reported in another investigation.²¹ Moreover, people tend to under- rather than over-report detrimental lifestyle habits such as smoking,³² thus any bias should tend to lower the retrieved associations. Although in the Fagerström score we did not exclude patients with missing answers, we attributed the lowest possible value to those answers; as a result, also in this case any bias should tend to lower the strength of the observed associations. Lastly, we did not collect data on physical activity.

In conclusion, with longer life expectancy due to better antiretroviral treatment, smoking has emerged as a frequent and life-threatening condition in HIV-infected patients, significantly influencing the life expectancy and quality of life. Italian HIV-infected patients demonstrated a high percentage of current smokers, ∼50% of cases, a large number of cigarettes smoked, and a remarkable proportion of people with high or very high degree of dependence. Our findings emphasize the need of smoking cessation strategies specifically targeting HIV-infected patients.

Authors' Contributions

The study concept and design were by G.V.D.S. Statistical expertise was contributed by E.R. Drafting of the article was carried out by G.V.D.S. and E.R. Analysis, interpretation of data, and critical revision of the article for important intellectual content were performed by all the authors. The authors had full access to the data and take responsibility for their integrity. All authors have read and agreed to the article as written.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding has been provided by any private nor public actor beyond the current clinical practice (provided by the Italian National Health Service).

Appendix

The STOPSHIV group comprises the following members. Coordination: G.V.D.S., E.R., and M.d.’O. Recruitment sites and investigators: Bari (C.S.); Busto Arsizio (B.M., T. Quirino, M. Farinazzo); Firenze (F.V.); Genova (A.D.B., F. Magne); Lecco (C. Molteni); Monza (N.S.); Perugia (G.V.D.S., E.S.); Sassari (G.M., P. Bagella, S. Mameli); Terni (D.F., B. Tiri); and Torino (G.O., M.G.).

References

Helleberg

, May

, Ingle

, et al.: Smoking and life expectancy among HIVinfected individuals on antiretroviral therapy in Europe and North America. AIDS, 2015; 29:221–229.

Sigel

, Wisnivesky

, Gordon

, et al.: HIV as an independent risk factor for incident lung cancer. AIDS, 2012; 26:1017–1025.

Reddy

, Kong

, Hyle

, et al.: Lung cancer mortality associated with smoking and smoking cessation among people living with HIV in the United States. JAMA Intern Med, 2017; 177:1613–1621.

Rasmussen

, Helleberg

, May

, et al.: Myocardial infarction among Danish HIV-infected individuals: Population-attributable fractions associated with smoking. Clin Infect Dis, 2015; 60:1415–1423.

European AIDS Clinical Society (EACS) guidelines, version 9.1. Web. 2018. Available at http://europeanaidsclinicalsociety.org (2018), accessed May 30, 2019 .

Pacek

, Latkin

, Crum

, et al.: Current cigarette smoking among HIV positive current and former drug users: Associations with individual and social characteristics. AIDS Behav, 2014; 18:1368–1377.

Mdodo

, Frazier

, Dube

, et al.: Cigarette smoking prevalence among adults with HIV compared with the general adult population in the United States: Cross-sectional surveys. Ann Intern Med, 2015; 162:335–344.

Bekele

, Rueda

, Gardner

, et al.: Trends and correlates of cigarette smoking and its impacts on health-related quality of life among people living with HIV: Findings from the Ontario HIV Treatment Network Cohort Study, 2008–2014. AIDS Patient Care STDS, 2017; 31:49–59.

Heatherton

, Kozlowski

, Frecker

, Fagerström

: The Fagerström test for nicotine dependence: A revision of the Fagerström Tolerance Questionnaire. Br J Addict, 1991; 86:1119–27.

10.

National Institute on Drug Abuse: Fagerström score. Available at https://cde.drugabuse.gov/instrument/d7c0b0f5-b865-e4de-e040-bb89ad43202b/module/f7cc1db9–2f13–70d5-e040-bb89ad4345a3/question/f7cc1db9- 2f15–70d5-e040-bb89ad4345a3 accessed May 30, 2019 .

11.

De Socio

, Ricci

, Maggi

, et al.: Prevalence, awareness, treatment, and control rate of hypertension in HIV-infected patients: the HIV-HY study. Am J Hypertens, 2014; 27:222–228.

12.

Rabe

, Hurd

, Anzueto

, et al.: Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Global Initiative for Chronic Obstructive Lung Disease. Am J Respir Crit Care Med, 2007; 176:532–555.

13.

Mancia

, Fagard

, Narkiewicz

, et al.: 2013 ESH/ESC Guidelines for the management of arterial hypertension: The Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens, 2013; 31:1281–1357.

14.

American Diabetes Association: Classification and diagnosis of diabetes. Diabetes Care, 2016; 39 (Suppl 1):S13–S22.

15.

Goff

Jr. , Lloyd-Jones

, Bennet

, et al.: 2013 ACC/AHA guideline on the assessment of cardiovascular risk: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines J Am Coll Cardiol 2014; 63: 2935–2959. Erratum in: J Am Coll Cardiol, 2014; 63 (25 Pt B):3026.

16.

Prignot

: Quantification and chemical markers of tobacco-exposure. Eur J Respir Dis, 1987; 70:1–7.

17.

ISTAT: [Health risk factors; smoking, obesity, alcohol, physical inactivity]. Fattori di rischio per la salute: fumo, obesità, alcol e sedentarietà.] tavole_allegate_fattori_rischio_salute_2016.xls Availabe at https://istat.it/it/archivio/202040 accessed May 30, 2019 .

18.

Weinberger

, Smith

, Funk

, et al.: Sex differences in tobacco use among persons living with HIV/AIDS: A systematic review and meta-analysis. J Acquir Immune Defic Syndr, 2017; 74:439–453.

19.

Tesoriero

, Gieryic

, Carrascal

, Lavigne

: Smoking among HIV positive New Yorkers: Prevalence, frequency, and opportunities for cessation. AIDS Behav, 2010; 14:824–835.

20.

Luo

, Duan

, et al.: Tobacco use among HIV-infected individuals in arural community in Yunnan Province, China. Drug Alcohol Depend, 2014; 134:144–150.

21.

Brath

, Grabovac

, Schalk

, et al.: Prevalence and correlates of smoking and readiness to quit smoking in people living with HIV in Austria and Germany. PLoS One, 2016; 11:e0150553.

22.

Feldman

, Minkoff

, Schneider

, et al.: Association of cigarette smoking with HIV prognosis among women in the HAART era: A report from the women's interagency HIV study. Am J Public Health, 2006; 96:1060–1065.

23.

King

, Vidrine

, Danysh

, et al.: Factors associated with non adherence to antiretroviral therapy in HIV-positive smokers. AIDS Patient Care STDS, 2012; 26:479–485.

24.

Nagao

, Hirayama

, Kon

, et al.: Current smokers with hyperlipidemia lack elevated preβ1-high-density lipoprotein concentrations. J Clin Lipidol, 2017; 11:242–249.

25.

Selya

, Hesse

: Time to first cigarette and serum cholesterol levels. Soc Sci Med, 2017; 174:213–219.

26.

Calvo

, Laguno

, Martínez

: Effects of tobacco smoking on HIV infected individuals. AIDS Rev, 2015; 17:47–55.

27.

Vijayaraghavan

, Penko

, Vittinghoff

, et al.: Smoking behaviours in a communitybased cohort of HIV-infected indigent adults. AIDS Behav, 2014; 18:535–43.

28.

De Socio

, Ricci

, Parruti

, et al.: Statins and Aspirin use in HIV-infected people: Gap between European AIDS Clinical Society guidelines and clinical practice: the results from HIV-HY study. Infection, 2016; 44:589–597.

29.

De Socio

, Pucci

, Baldelli

, Schillaci

: Observed versus predicted cardiovascular events and all-cause death in HIV infection: A longitudinal cohort study. BMC Infect Dis, 2017; 17:414.

30.

[Italian guidelines on antiretroviral therapy use and diagnostic and clinical management of HIV patients]. Linee Guida Italiane sull'utilizzo della Terapia Antiretrovirale e la gestione diagnostico-clinica delle persone con infezione da HIV-1]. Edizione 2017. Available at http://salute.gov.it/imgs/C_17_pubblicazioni_2696_allegato.pdf accessed May 30, 2019 .

31.

EpiCentro: [HIV infection and AIDS. Epidemiological aspects]. Infezione da HIV e AIDS. Aspetti epidemiologici. Available at https://epicentro.iss.it/aids/epidemiologia-italia accessed April, 2019.

32.

Gallus

, Tramacere

, Boffetta

, et al.: Temporal changes of under-reporting of cigarette consumption in population-based studies. Tob Control, 2011; 20:34–39.