Daily Cannabis Use Is a Barrier to Tobacco Cessation Among Tobacco Quitline Callers at 7-Month Follow-up

Abstract

Introduction

Cannabis use is increasing among cigarette smokers in the United States. Prior studies suggest that cannabis use may be a barrier to smoking cessation. Yet, the extent to which this is the case among adults seeking to quit tobacco use remains unclear. Tobacco quitlines are the most common provider of no-cost treatment for adults who use smoke in the United States. This study investigated the association between cannabis use and smoking cessation outcomes among quitline callers.

Aims and Methods

Participants included callers to the New York State Smokers’ Quitline, who were seeking to quit smoking cigarettes and were contacted for outcome assessment 7 months after intake. Thirty-day point prevalence abstinence rates were calculated and compared among cannabis use groups, based on frequency of past-30-day cannabis use at baseline (none: 0 days, occasional: 1–9 days, regular: 10–19 days, and daily: 20–30 days).

Results

Approximately 8.3% (n = 283) of participants (n = 3396) reported past-30-day cannabis use at baseline. Callers with daily cannabis use (20–30 days per month) had significantly lower odds of 30-day abstinence, relative to those who did not use cannabis (odds ratio = 0.5; 95% confidence interval [0.3, 0.9]).

Conclusions

Daily cannabis use appears to be associated with poorer smoking cessation treatment outcomes among adults seeking to quit smoking cigarettes via a quitline. Because quitlines are among the most accessible, affordable, and frequently utilized community-based treatments available in the United States, and the prevalence of cannabis use is increasing among cigarette smokers, detailed inquiry into cannabis use might enhance cigarette smoking cessation outcomes.

Implications

Quitlines are free of cost and accessible to millions of smokers in the United States. The current study found an inverse relationship between daily cannabis use at baseline and 30-day abstinence from cigarette smoking at 7-month follow-up among New York State Smokers’ Quitline callers. Findings suggest that daily cannabis use may be a barrier to smoking cessation and sustained abstinence among those seeking help to stop smoking cigarettes.

Introduction

Cigarette smoking remains the leading cause of preventable disease and premature death in the United States.^1,2 Smoking cessation significantly reduces tobacco-related disease.³ Evidence-based treatment for tobacco dependence increases the odds of achieving abstinence,⁴ but not all groups benefit equally. Prior data are mixed, with some studies suggesting that cannabis use is associated with lower likelihood of smoking cessation and increased odds of relapse to cigarette use, and others finding no link.^5–11 Increases in the use of cannabis and recent epidemiological evidence that cannabis use is associated with persistent smoking has given rise to questions about the impact of cannabis use on evidence-based smoking cessation treatment outcomes.

Cannabis and tobacco use are intertwined. Nationally, approximately 9% of cigarette smokers report past-30-day daily cannabis use, relative to 1% of nonsmokers, and the prevalence of cannabis use has increased in the past decade among both smokers and nonsmokers.^12,13 Further, over two-thirds of all US residents now live in a state where cannabis is legal in some form,¹⁴ and cannabis use is increasing even more rapidly in states where recreational use has been legalized.^15,16 Epidemiologic data suggest that cannabis use is associated with persistence of cigarette use, nicotine dependence, lower quit rates, and relapse to cigarette smoking among former smokers in the United States.^7,17,18

Free, accessible, and evidence-based treatment is available by telephone in every state in the United States via tobacco quitlines, making quitlines the largest tobacco treatment network in the United States.^19,20 At present, there is only one previous study assessing cannabis use and smoking cessation outcomes among quitline callers. In this study, cannabis use was assessed as a categorical variable (yes/no) via electronic medical record review.²¹ No association was found between cannabis use and abstinence from smoking. However, these findings might be a result of the binary categorization of cannabis use. The nature of the relations between smoking cessation and cannabis use might depend on the frequency of cannabis use, which has not been measured in prior quitline studies. Mental health factors could also confound the relationship between cannabis use and smoking session given that cannabis use is associated with increased mental health problems,^22–24 and mental health problems are associated with poorer smoking cessation treatment outcomes.^25,26

The current study examined the frequency of cannabis use among callers to the New York State Smokers’ Quitline (NYSSQL), and the degree to which frequency of past-30-day cannabis use at intake was associated with 30-day point prevalence cigarette abstinence outcomes during the 7-month outcome assessment adjusting for potential confounders (eg, number of cigarettes smoked per day, alcohol use, and psychological distress). At the time of data collection, recreational cannabis was prohibited by law in New York State.

Methods

Study Population

This study was approved by the Roswell Park Comprehensive Cancer Center Institutional Review Board. Data were extracted from a dataset compiled as part of program evaluation activities for the NYSSQL. The sample comprised all callers who smoked cigarettes, provided a response to the cannabis use assessment during intake from May 2018 to June 2020, and responded to the 7-month outcome assessment.

Procedures

All callers were administered the standard intake assessment, screened for nicotine replacement therapy eligibility, and provided a brief counseling session by a Quitline Coach during their first call. As part of standard quality improvement processes, a subset of callers was routinely contacted and invited to complete an outcome assessment 7 months after intake. All callers with email addresses were invited to complete the assessment online. A random sample of n = 200 callers per week who do not have email addresses were invited to complete the assessment by telephone. Finally, a random sample of n = 50 callers with email addresses who do not respond to the email invitations after 5 reminders were contacted by telephone and invited to complete the assessment. Specially trained interviewers conducted the outcome assessments over the telephone.

Measures

Sociodemographic assessment included sex, age, race/ethnicity, education, and insurance status. Baseline tobacco use and clinical measures included number of cigarettes smoked per day, use of other tobacco products, years smoking cigarettes, readiness to quit, living with others who smoke cigarettes, and importance of and confidence in quitting cigarettes. Past-30-day alcohol use was assessed with the questions: “How many days in the past month did you drink alcohol?” If the caller reported >0, they were asked, “How many days in the past month did you have 5 or more drinks in one sitting?” and “On a scale of 0 to 10 where 0 = not at all and 10 = most possible, how much do you want to reduce your alcohol consumption?”²⁷ Past-30-day psychological distress was assessed with the Kessler Psychological Distress Scale (K6), which assesses the frequency of past-30-day feelings of nervousness, hopelessness, restlessness, sadness, or depression, feeling that everything was an effort, and feelings of worthlessness.²⁸ A K6 score of ≥5 and <13 is indicative of moderate psychological distress²⁹; and ≥13 is indicative of severe psychological distress.^30,31

Cannabis Use

Cannabis use was assessed by asking, “How many days of the past month have you used any form of cannabis or marijuana? This includes weed, pot, herb, grass, edibles, kief, wax, glass, shatter, hashish, hash oil, or butane honey oil.” Please respond “0” if you only used CBD products. Cannabis use was categorized as a four-level variable based on the frequency of use—none: 0 days, occasional: 1–9 days, regular: 10–19 days, and daily (encompassing almost daily use): 20–30 days.

Outcome Measure

Point prevalence abstinence was assessed 7 months after intake with the questions, “How many cigarettes are you smoking on a usual day?” followed by, “Have you smoked a cigarette, even a puff, in the last 30 days?”

Data Analysis

Analysis of variance and chi-squared analyses were used to examine differences between callers who did and did not report any cannabis use at intake. Complete case analysis was used to calculate abstinence rates (ie, only those callers who responded to the outcome assessment were included). To explicate the nature of the association between cannabis use and smoking cessation, cannabis use at intake as the exposure variable was analyzed as a four-level categorical variable. Adjusted and unadjusted odds ratios with 95% confidence intervals were estimated. Analyses were conducted with SPSS (Version 25.0).

Results

Participants

Among all unique quitline callers to the NYSSQL from May 2018 to June 2020 (n = 55 821), 81.33% (n = 45 398) provided frequency of past-30-day cannabis use and 18.67% responded “don’t know” or refused to answer the question. Of those who reported frequency of past-30-day cannabis use, 61.41% (n = 27 878) were invited to complete the outcome assessment. Of those invited to complete the outcome assessment, 12.18% (n = 3396) provided 30-day point prevalence data at 7-month follow-up.

Significant differences between participants who did (n = 283) and did not use cannabis (n = 3113) were few. Of the 283 participants who reported cannabis use, 137 reported occasional use (1–9 days), 37 reported regular use (11–20 days), and 109 reported daily/almost daily use (20–30 days). Cannabis use was more common among men, those who live with other cigarette smokers, those who use other non-cigarette tobacco products and those with higher levels of serious psychological distress, compared with those who did not (see Table 1).

Table 1.

Factors Associated With Cannabis Use Among New York State Quitline Callers (N = 3402)

Variable	Range or categories	Percent (n) or mean (SD)		p
		No cannabis use (n = 3113)	Cannabis use (n = 283)
Demographic differences at baseline
Sex	Female	55.2% (1717)	48.8% (138)	.039
Age (mean)	19–90	57.35 (13.30)	51.34 (14.06)	.001
Race/ethnicity	White	55.1% (1715)	55.1% (156)	.426
	Black	18.8% (585)	22.3% (63)
	Hispanic	12.9% (401)	13.8% (39)
	Other	3.9% (120)	2.5% (7)
Education	High school or less	44.7% (1392)	48.1% (136)	.059
	Some college/trade school	21.6% (672)	29.7% (84)
	College degree	16.2% (503)	14.1% (40)
Insurance	Yes	66.3% (1878)	71.5% (186)	.088
Medicaid/Medicare	Medicaid	25.4% (718)	37.7% (98)	.001
	Medicare	28.3% (800)	21.5% (56)
	Both	12.7% (360)	12.3% (32)
	Neither	33.7% (953)	28.5% (74)
Features of cigarette use
Cigarettes per day (mean)	0–80	17.71 (10.33)	17.08 (10.38)	.324
Years smoking cigarettes (mean)	0–71	32.30 (15.42)	30.30 (14.72)	.047
Ready to quit in the next 30 days	Yes	87.8% (2734)	89.4% (253)	.543
Live with other cigarette smokers	Yes	10.0% (310)	16.3% (46)	.012
Use other tobacco products in past 30 days	Yes	2.2% (67)	3.9% (11)	.072
Tried to quit before	Yes	85.8% (2672)	83.4% (236)	.262
Mental health and other substance use
K6 (mean)	0–24	5.16 (5.10)	7.11 (5.59)	.001
Past-30-day alcohol use (mean)	0–30 days	1.80 (5.47)	4.84 (8.45)	.001
Past-30-daybinge drinking (mean)	0–30 days	1.71 (5.30)	2.45 (6.00)	.154

K6 = Kessler Psychological Distress Scale.

Thirty-Day Point Prevalence Abstinence From Smoking Cigarettes

Daily cannabis use (20–30 days per month) was associated with significantly lower point prevalence 30-day abstinence rates (odds ratio = 0.544 [0.330, 0.898]; p = .017), relative to no cannabis use. These results remained unchanged after adjusting for demographics (ie, age, sex, race/ethnicity, and education) and psychological distress (see Table 2). Occasional cannabis use (1–9 days) was associated with significantly lower point prevalence 30-day abstinence rates (odds ratio = 0.514 [0.266, 0.993]; p = .048), relative to no cannabis use, after adjusting for psychological distress only.

Table 2.

Past-30-Day Cannabis Use and Past 30-Day Abstinence From Cigarette Use at 7-Month Follow-up

		p	OR	95% CI
				Lower	Upper
Days of cannabis usea,d	1–9	.084	0.692	0.456	1.051
	10–19	.901	0.955	0.460	1.981
	20–30	.017	0.544	0.330	0.898
Days of cannabis useb,d	1–9	.080	0.675	0.435	1.048
	10–19	.919	1.042	0.472	2.300
	20–30	.018	0.523	0.305	0.897
Days of cannabis usec,d	1–9	.048	0.514	0.266	0.993
	10–19	.107	0.188	0.025	1.438
	20–30	.040	0.471	0.229	0.968

CI = confidence interval, OR = odds ratio.

Bold values reflect p < .05.

Unadjusted.

Adjusted for age, sex, race/ethnicity, and education.

Adjusted for serious psychological distress only.

Compared with no cannabis use in the past 30 days.

Discussion

These findings suggest that daily cannabis use may be associated with poorer smoking cessation outcomes at 7 months among tobacco quitline callers. The relationship between daily cannabis use and abstinence rates 7 months later among quitline callers does not appear to be due to confounding by demographics or mental health. Our findings extend extant literature suggesting cannabis may be a barrier to successful long-term cigarette abstinence, showing this effect at 7 months follow-up and among a “real world” sample of quitline callers in the community.⁹ Differences in measurement, exclusion criteria, and follow-up period may explain mixed findings in this topic area overall.^5–8,10,11

With cannabis use increasing across the United States and worldwide, and data showing cannabis use is much more common among individuals who smoke cigarettes, it is reasonable to predict that the proportion of callers to tobacco quitlines who use cannabis may increase over the coming years. These changes in the substance use landscape, in combination with the findings of the current study, highlight the need for surveillance and capture of cannabis use as part of quitline screening.³² Lack of measurement of the amount of cannabis used per day and the mode of administration, including mixing with tobacco, in the current study limit these findings in terms of understanding how these factors affect smoking cessation. Cannabis is also an umbrella term for multiple products, for example, CBD-only versus THC-dominant products, that may result in different psychoactive effects. Capturing detailed cannabis information in future studies, including biochemical verification to confirm self-reported usage amounts, may be challenging; however, measures of severity of use/amount of exposure will be critical to informing intervention efforts.³³ Additional data on reasons for cannabis use and social and environmental factors associated with use may also impact the degree to which cannabis use affects tobacco cessation and may inform how best to intervene.³²

Another limitation of the study is that complete case analyses incorporates only those participants who complete outcome assessments and is often assumed to be positively biased. This concern is lessoned by the assumption that using complete case analyses would affect those who did and did not use cannabis equally. Baseline characteristics between those who responded to the 7-month follow-up assessment and those who did not respond differed significantly, but these differences are occur equally across cannabis and non-cannabis users and are therefore unlikely to significantly affect study outcomes. Use of data from a community-based, publicly funded quitline (vs. a highly controlled treatment study) facilitates generalizability though response rates at 7-month follow-up are low due in part to lack of funding, suggesting replication is needed. Replication should also be conducted in different states due to diverse features (eg, dispensary density) and implementation of cannabis laws. Finally, future studies that prospectively collect information on both cannabis and tobacco use among quitline callers at follow-up are needed next as patterns of cannabis use may change during tobacco treatment, which could be relevant to understanding quitline outcomes, particularly if substitution of cannabis is occurring with tobacco reduction or cessation.

Our findings suggest that daily cannabis use has an adverse impact on tobacco cessation among those actively trying to quit with treatment and highlights the relevance of measuring cannabis use in detail (ie, the amount and frequency). A negative relationship between occasional cannabis use (1–9 days) and abstinence at follow-up also emerged when adjusting for mental health problems and does not appear to be an artifact of sample size. A yes/no cannabis use response may not provide enough detailed or clinically meaningful information to guide next steps in research, prevention, and intervention. Our findings, which are largely consistent with prior results, suggest that routine assessment of cannabis use is advisable among quitline treatment providers, the largest network of tobacco treatment services in the United States.

Supplementary Material

A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr.

Funding

P30DA029926 (Budney).

Declaration of Interests

None declared.

Data Availability

Due to ethical considerations, quitline data cannot be shared as these data are primarily used for clinical purposes through a contract from the NYS Department of Health to Roswell Park Comprehensive Cancer Center for the purpose of providing tobacco cessation assistance.