E-Cigarette Use and Depression Among American Indian Adults Who Smoke

Abstract

Objective

This study examined the association of e-cigarette use status and history of depression among American Indian (AI) adults who smoke.

Method

We conducted a secondary data analysis using survey data from 375 AI adult smokers collected in 2016 at a tribally operated healthcare facility in northeast Oklahoma. Multivariable logistic regression was used to estimate the association between e-cigarette use and self-reported history of depression while adjusting for potential confounders.

Results

In the adjusted analyses, compared to never users, current and former e-cigarette users had higher estimated odds of depression history (adj. OR 2.66; 95% CI 1.25–5.72 and adj. OR 2.38; 95% CI 1.36–4.26, respectively). Additional factors independently associated with a history of depression included having strong cravings to smoke (adj. OR 2.28; 95% CI 1.13–4.88) and having a history of chronic disease (adj. OR 2.09; 95% CI 1.20–3.70) after controlling for confounding variables.

Conclusions

E-cigarette use among AI adult smokers was independently associated with a history of depression. Whether e-cigarette use among people who smoke results from depression or whether depression results from the combined use of e-cigarettes and cigarettes requires future research using a prospective cohort design.

American Indians (AIs) suffer disproportionally from mental health conditions and chronic diseases related to tobacco use [1]. Despite overall decreases in United States (US) smoking rates over the past 50 years, nearly one in five AIs currently smoke, which is nearly twice the national prevalence [2]. Consequently, AIs experience smoking related cancers (e.g., lung cancer) and smoking related mortality at disproportionately high rates [3–7]. In addition, data from the 2018 Behavioral Risk Factor Surveillance System (BRFSS) indicated that 40% of AI adults self-reported poor mental health, which was highest compared to all other ethnic/racial groups [8]. In a nationally representative study, Hasin and colleagues (2018) found that AIs had the highest 12-month (15.9%) and lifetime prevalence (28.2%) of major depressive disorder (MDD) compared to non-Hispanic Whites (NHWs) [9]. In 2019, nearly 30% of AIs in Oklahoma reported receiving a diagnosis of some form of depression by a medical provider compared to 23% for NHW adults and 19% for Black adults [10].

Epidemiological evidence has established strong associations between mental health disorders and regular smoking for AIs [11, 12]. Mental health and substance use disorders (SUDs) among AIs frequently co-occur [13]. According to an Indian Health Service (IHS) report, MDD and tobacco use disorder were among the top co-occurring conditions for outpatient behavioral health treatment for AIs 18 years of age and older [14]. In one study, the lifetime prevalence of mood disorders was roughly double for AIs with co-occurring nicotine dependence compared to AIs without nicotine dependence [15]. Another study indicated that southwest AI adults with depression had two times higher odds of smoking compared to a northern plains AI sample [16]. Collectively, both AI groups with a self-reported mental health disorder had higher odds of smoking.

Electronic cigarettes (e-cigarettes) are changing the landscape of tobacco and nicotine product use. E-cigarettes have been marketed as less harmful alternatives to combustible cigarettes and have become increasingly popular among adults who smoke and who want to reduce or quit cigarette smoking [17]. Studies show e-cigarettes may be an effective form of smoking cessation compared to nicotine replacement therapy (NRT) [18]. However, it is not an FDA-approved method, nor is it generally recommended. [18–22]. Concomitant e-cigarette and cigarette use, or dual use, prevalence among AIs is high and varies by geographic region and by sociodemographic factors, such as employment and education [23].

Although research on e-cigarette use and depression is still emerging, there is growing evidence supporting an e-cigarette use and depression connection, aligning with the self-medication hypothesis [24, 25]. The self-medication hypothesis suggests that individuals might turn to substances like tobacco or e-cigarettes to cope with psychiatric symptoms. This theory highlights how substance use can serve as an attempt to alleviate symptoms of mental health challenges such as stress, anxiety, or depression [25–27]. AI’s and other vulnerable populations may be particularly prone to using e-cigarettes as a coping mechanism due to persistent socioeconomic disparities, chronic stress, and generational trauma [28, 29].

Moreover, BRFSS data suggest that individuals who currently use e-cigarettes have twice the risk of depression symptoms compared to non-users [10]. In one study, e-cigarette use was associated with depression among those who reported marijuana use, were unemployed, and were not married. Another study found former users of e-cigarettes had 60% higher odds of depression, while current e-cigarette use was associated with twice the odds of depression than non-use [30]. Kang and Malvaso [31] utilized a longitudinal dataset from the UK to examine the relationship between e-cigarette use and various aspects of mental health, including depression. While significant associations were found between e-cigarette use and adverse general mental health factors, no direct link to depression was observed. However, a notable association emerged between e-cigarette use and anhedonia, a specific depressive symptom marked by a loss of interest or pleasure in previously enjoyable activities.

In general, adverse behavioral health effects of e-cigarette use among individuals who smoke (dual use) is an under-researched area, with even fewer studies focused on these factors across AI populations. One study found that African Americans were more likely to maintain dual use than Hispanic and NHW sub-groups; however, the study focused more on the intentions/reasons for dual use rather than the potential relationship between dual use and health outcomes [32]. In another study, dual use was associated with a much higher risk of moderate to severe depressive symptoms than individuals who used only one product in South Korea [33].

AI adults have among the highest prevalence in the US of smoking, e-cigarette use, and depression. Yet, studies identifying the association between depression and e-cigarette use among AI smokers are extremely sparse. The lack of research in this area may contribute to the widening health disparities in this population. The current paper aimed to conduct a secondary data analysis to identify the association between e-cigarette use and history of depression among AI adult smokers.

Method

Study Design

This cross-sectional study examined baseline survey data collected in 2016 from a study of AI adults and their use of cigarettes. The survey included questions measuring sociodemographics, e-cigarette use, depression, and other factors. The current study is a secondary analysis of these data, which was conducted to directly explore the relationship between e-cigarette use status and depression. For recruitment methods, research staff recruited a convenience sample of participants from a high-traffic waiting area within one of Cherokee Nation Health System’s primary care outpatient facilities. On recruitment days, research staff set up a table with information about the study. Unless approached first, research staff approached potential participants and invited them to hear more about the study. Participants were eligible for the study if they currently smoked cigarettes (any use in the past 30 days, but smoked more than 100 cigarettes in their lifetime), were 18 years or older, and were of AI descent. Besides completing a survey, study participants also provided a carbon monoxide (CO) sample and a saliva sample to check for CO and cotinine levels. A total of (n = 375) adult AI individuals were provided informed consent and were enrolled in the study. For additional information about the initial study, see Comiford and colleagues [34]. The Cherokee Nation Institutional Review Board (IRB) (IRB #274) and the University of Oklahoma Health Sciences Center IRB (IRB #5970) approved the initial study. The current paper was approved for publication consideration by the Cherokee Nation IRB, with additional approvals obtained from the Cherokee Nation Communications Department and the Office of the Cherokee Nation Chief of Staff.

E-Cigarette Use

E-cigarette use was measured by two items. The first item (Yes/No) asked: “Have you ever vaped or used an e-cig even one or two times?” The second item asked: “On how many days of the past 30 days did you use an e-cig or vape even one or two times?” with response options of “0,” “1–10 days,” “11–20 days,” and “21–30 days.” This item was treated as “0” if participants answered “No” to the first. Responses were coded into three groups: “Current Users,” “Former Users,” and “Never Users.” Participants were considered current e-cigarette users if they reported any use within the past 30 days. Participants were considered former users if they reported ever using an e-cigarette, but no use in the past 30 days. Lastly, participants were considered never users if they reported never having used an e-cigarette.

Depression

Due to concerns about respondent burden, depression history was assessed with a single item adapted from the BRFSS [35]. Depression, as a dependent variable, was treated as a binary outcome (No = 0, Yes = 1), which was assessed as follows: “Place a checkmark in the column to indicate if a doctor or health care professional has told you that you had depression.” Accessing medical record data or conducting detailed depression screenings were not within the scope of the original study.

Covariates

The following sociodemographic characteristics were included as covariates based on prior research demonstrating associations with cigarette cessation and use of e-cigarettes: age, sex, annual income, and education. Other characteristics included as covariates were smoking cravings, chronic disease history, smoking cravings, e-cigarette harm perceptions, and confidence to stop using all tobacco products, e-cigarette harm perceptions, and confidence to stop using all tobacco products [35, 36]. Age was grouped into four categories (18–25, 26–35, 36–45, and 45 +). Sex was dichotomized as male or female. Annual household income was dichotomized into $0–$30,000 and $30,001 + categories. Education was defined as the highest level of school completed and was grouped into “less than high school,” “high school/GED,” and “more than high school” categories. Chronic disease history was defined as the participant’s self-reported history of having one or more of the following conditions: cancer, cardiovascular disease, diabetes, chronic obstructive pulmonary disease, or emphysema; all of which were presented as dichotomous (“Yes” vs. “No”) questions in the survey. Response options to the strong cravings to smoke item were “Yes” vs. “No,” while responses to the “e-cigs are less harmful than cigarettes” item were categorized as “Yes” vs. “No/Don’t Know/Not Sure.” Confidence to quit smoking was measured in the original survey as a 10-point Likert-scale question asking, “How confident are you that you will quit smoking within the next month?” and then dichotomized into “high confidence” vs. “low,” with high confidence being defined as a rating of 6 out of 10 or higher.

Statistical Analyses

R (version 3.5.1) was used for data analyses. Factors potentially associated with history of depression were assessed using chi-square tests for independence or Fisher’s exact test when warranted by cell counts. Significant factors were retained as adjustment terms for multivariable logistic regression models if the p-value for the univariate association was < 0.05. Crude odds ratios (ORs) were also estimated from logistic regression models. Adjusted odds ratios (adj. ORs) were estimated from logistic regression models including vaping status, cravings, sex, age, income, and chronic health history. In both cases, term-specific p-values were computed using Wald chi-square tests. A combined, final model was derived from a saturated model (including pre-specified interactions between vaping status and sex, age, income, and education) using backwards selection, first by removing non-significant interaction terms, and then main effects unless their removal resulted in a change of 20% or greater in other coefficients; in this case, they were retained as adjustment terms. All tests were two-sided with statistical significance defined as p-value < 0.05.

Results

Table 1 presents characteristics of the overall study sample by depression status. The majority of participants were female (62%), and the average age of participants was 41.7 (SD = 13.3) years. Vaping status was significantly associated with history of depression χ² (1, N = 358) = 13.36, p = 0.001. Only 29% of never users of e-cigarettes reported a history of depression compared to 49% of former users of e-cigarettes, and 48% of current users of e-cigarettes, respectively. Factors significantly associated with history of depression included female sex χ² (1, N = 358) = 5.63, p = 0.02, lower income χ² (1, N = 350) = 4.89, p = 0.027, strong cravings to smoke χ² (1, N = 358) = 12.39, p ≤ 0.001, and a personal history of chronic disease χ² (1, N = 358) = 12.99, p ≤ 0.001. Factors that were not significantly associated with history of depression included age, education, perception of e-cigarettes as being less harmful than cigarettes, and confidence to stop using all tobacco products.

Table 1.

Sample characteristics by depression status among AI adults who smoke

Characteristic	Overall(1) (n = 358)	Yes(2) (n = 148)	No(2) (n = 210)	p-value(3)
Vape usage status				.001
Never	129 (36.03%)	37 (28.68%)	92 (71.32%)
Past	170 (47.49%)	83 (48.82%)	87 (51.18%)
Current	59 (16.48%)	28 (47.46%)	31 (52.54%)
Sex				.018
Female	222 (62.01%)	103 (46.40%)	119 (53.60%)
Male	136 (37.99%)	45 (33.09%)	91 (66.91%)
Age				.664
18–25	48 (13.41%)	16 (33.33%)	32 (66.67%)
26–35	90 (25.14%)	37 (41.11%)	53 (58.89%)
36–45	76 (21.23%)	33 (43.42%)	43 (56.58%)
46 +	144 (40.22%)	62 (43.06%)	82 (56.94%)
Age in years				.499(4)
Mean ± SD	41.66 ± 13.31	42.23 ± 12.77	41.26 ± 13.69
Median [25%, 75%]	41 (31, 53)	42 (31, 52)	39 (30, 53)
Range	18, 77	19, 77	18, 75
N missing	0	0	0
Education				.818
Less than high school	72 (20.17%)	32 (44.44%)	40 (55.56%)
High school/GED	137 (38.38%)	55 (40.15%)	82 (59.85%)
More than high school	148 (41.46%)	60 (40.54%)	88 (59.46%)
Income				.027
$0–$30,000	275 (78.57%)	122 (44.36%)	153 (55.64%)
$30,000 +	75 (21.43%)	22 (29.33%)	53 (70.67%)
Strong cravings to smoke				< .001
No	71 (20.11%)	16 (22.54%)	55 (77.46%)
Yes	282 (79.89%)	131 (46.45%)	151 (53.55%)
Chronic disease history				< .001
No	176 (56.41%)	54 (30.68%)	122 (69.32%)
Yes	136 (43.59%)	70 (51.47%)	66 (48.53%)
E-cigs less harmful than cigs				.163
No/DK/NS	242 (68.17%)	93 (38.43%)	149 (61.57%)
Yes	113 (31.83%)	53 (46.90%)	60 (53.10%)
Confidence to stop tobacco				.361
Low	306 (86.20%)	131 (42.81%)	175 (57.19%)
High	49 (13.80%)	17 (34.69%)	32 (65.31%)

⁽¹⁾Percentages computed within total of characteristic

⁽²⁾Percentages computed within level of characteristic

⁽³⁾p-values from chi-square test for independence or Fisher’s exact test, where warranted by low expected cell counts

⁽⁴⁾p-value for continuous age in years computed via two-sample t-test

Table 2 presents findings from the unadjusted and adjusted odds ratios (ORs), confidence intervals (CI), and p-values for the logistic regression models. In the final multivariable model, vaping status remained independently associated with a history of depression. Participants who formerly used e-cigarette had 2.4 times greater odds of depression, and current e-cigarette use was associated with 2.7 times greater odds of depression compared to AI participants who never used e-cigarettes. Having strong cravings to smoke was associated with 2.3 times higher odds of depression compared to those who did not report strong cravings to smoke. Individuals with a chronic disease history had twice the odds of depression compared to those without a chronic disease history. No significant interactions with vaping status were found for age, sex, income, and education.

Table 2.

Multiple logistic regression results for predictors of depression, unadjusted, and adjusting for vaping status, sex, age group, income, strong cravings, and chronic disease history

Predictor	Crude				Adjusted(1)
	OR	LCL	UCL	p-value	OR	LCL	UCL	p-value
Vape usage status				0.0007				< .001
Never	(Ref)				(Ref)
Past	2.37	1.47	3.88		2.38	1.36	4.26
Current	2.25	1.19	4.27		2.66	1.25	5.72
Sex				0.0135				.088
Female	(Ref)				(Ref)
Male	0.57	0.36	0.89		0.64	0.38	1.06
Age				0.2329				.484
18–25	(Ref)				(Ref)
26–35	1.40	0.68	2.95		1.88	0.75	4.99
36–45	1.53	0.73	3.30		2.06	0.81	5.56
46 +	1.51	0.77	3.06		1.95	0.79	5.10
Education				0.5278				-
Less than high school	(Ref)
High school/GED	0.84	0.47	1.50
More than high school	0.85	0.48	1.51
Income				0.0202				.051
$0–$30,000	(Ref)				(Ref)
$30,000 +	0.52	0.30	0.89		0.53	0.28	0.98
Strong cravings to smoke				0.0004				.025
No	(Ref)				(Ref)
Yes	2.98	1.66	5.61		2.28	1.13	4.88
Chronic disease history				0.0002				.010
No	(Ref)				(Ref)
Yes	2.40	1.51	3.83		2.09	1.20	3.70
E-cigs less harmful than cigs				0.1314				-
No/DK/NS	(Ref)
Yes	1.42	0.90	2.22
Confidence to stop tobacco				0.2863				-
Low	(Ref)
High	0.71	0.37	1.32

OR odds ratio, LCL lower 95% confidence limit, UCL upper 95% confidence limit

⁽¹⁾Adjusted model includes terms for: vape usage group, gender, income, strong cravings, and personal medical history

Discussion

This study builds on previous research examining e-cigarette use among AI individuals who smoke by directly investigating the relationship between e-cigarette use and history of depression in sample of AI adults who smoke. Approximately half of participants who formerly and currently used e-cigarettes reported history of depression while 30% of individuals who never used e-cigarettes reported a history of depression. Significant bivariate associations were observed with depression, e-cigarette usage, female sex, income, strong cravings to smoke, and chronic disease history; however, the associations with sex and income did not hold when adjusted for other predictors in the multivariable analyses.

Our findings are consistent with the general literature on the relation between e-cigarettes and depression and with prior literature demonstrating relations between dual e-cigarette use and smoking among AI samples [5, 15, 33, 34]. The current findings showed that the relationship between e-cigarette use and depression remained even after adjustment for sociodemographic and other factors. Our findings also revealed important associations with depression and strong cravings to smoke, as well as self-reported chronic disease history.

E-cigarette use is highest among persons who currently smoke and these individuals often vape to reduce or quit smoking, including in our sample of AI adults [36, 37]. In this study, people who smoke and who formerly or currently used e-cigarettes were more likely to report depression history compared to people who smoke but never used e-cigarettes.

Our findings indicated that former or current dual use of e-cigarettes and cigarette smoking was positively associated with depression among AI adults. In the general population, higher odds of depression have been observed among dual users compared to non-users [38]. Other studies on tobacco cessation and depression have indicated reductions in depression symptoms following a successful smoking quit attempt.

The effect of increased exposure to nicotine and possible nicotine dependence from dual e-cigarette and cigarette use might be one way to interpret the current results. Increased nicotine exposure levels increase the risk of nicotine dependence, undermine quit attempts, and can lower serotonin levels, which impact increased psychological distress, such as depressive symptoms [39]. Excessive nicotine exposure from dual use has been found to disrupt the development of neural pathways in areas of the brain responsible for mood regulation, stress sensitivity, and adaptive coping ability [40]. Depressive symptoms are a known barrier to smoking cessation [41, 42]. It is also possible that smokers who try to quit but are unable to do so experience depression, a problem that may be more acute for persons with chronic disease. It is possible that healthy individuals (i.e., those without chronic disease and depression) may have fewer motivations to quit smoking and may be less likely to try or use e-cigarettes as e-cigarettes are often used to reduce or quit cigarette smoking. Another potential reason for this association is that persons experiencing negative mood symptoms are using e-cigarettes and/or cigarettes to manage/cope with their negative emotions, which is in line with the self-medication theory [25, 26]. In our sample, participants reporting strong cravings to smoke had a moderate to strong association with depression compared to those who did not have strong cravings to smoke. Perhaps the connection between strong cravings and depression is reflective of the need to smoke or vape to cope with one’s depressive feelings. However, this relationship should be investigated in future studies [43, 44].

The current findings should be considered within the context of the study’s limitations. First, we used a single self-report measure of ever being diagnosed with depression, as depression was not the focus of the original study. As such, we did not have the ability to identify depression sub-types, specific diagnoses, or the severity of depressive symptoms among the participants. The use of a depression screener in combination with access to medical record information may allow future research to examine e-cigarette use status in relation to depression in richer detail. Second, due to the cross-sectional design of the parent study, it cannot be determined if the participants experienced depression prior to or following their initiation of e-cigarette use or smoking. Longitudinal studies are needed to identify the temporal relationship between dual use and depression. Third, this study relied on self-report data, which introduces the possibility of recall bias. Fourth, the sex distribution in the sample was approximately 60% female and 40% male. As reported in other studies, depression in this study was reported by women more often than for men. Although confounding by sex cannot be completely excluded, it is unlikely based on the lack of interaction of vaping with sex as well as adjusting for sex in the multivariate models. Tobacco use rates are also known to vary by sex and geographic region among AI groups [45]. Thirty-eight percent of AI females in northeast Oklahoma self-reported as currently using cigarettes, which was higher than AI males (34%) as well as NHW females (19%) [4]. Participants were recruited using convenience sampling from a tribal outpatient health clinic in northeast Oklahoma. Thus, our results may not generalize to other AI/AN populations located in different regions and geographical settings (e.g., urban, rural, reservation/reserve). Lastly, although these findings are relevant, research using newer data sources may be warranted given that e-cigarette products and usage patterns have changed since 2016.

Despite these limitations, our study has important implications. The findings support community health actions to improve awareness of the complex relationship between depression, e-cigarette use, and commercial tobacco use. For example, these results could improve the implementation of culturally adapted tobacco cessation interventions in AI healthcare facilities by incorporating routine depression screenings and referrals to ancillary services, such as behavioral health, as part of a comprehensive tobacco cessation approach. Screening, brief intervention, and referral to treatment (SBIRT) has proven effective with smoking cessation and has demonstrated promise at reducing substance use risk (including tobacco use initiation) among rural dwelling AI youth [46, 47]. In addition, tobacco cessation interventions that are evidence-based and incorporate cultural practices and values (e.g., understanding the use of traditional/ceremonial tobacco versus commercial tobacco) offer a full spectrum of care that is effective for substance use treatment among AI adults, and yet, is often absent from AI serving health facilities. Understanding the extent to which e-cigarettes are being used by this population to self-medicate depressive symptoms will also be important to investigate further. The risk of developing new onset substance use disorders among those who self-medicate substantially increases over time compared to those with mood disorders who do not self-medicate [26]. Thus, early screening and intervention implementation is imperative. Lastly, these findings could be used as preliminary data for future in-depth research on the association between depression and e-cigarette use.

Conclusions

This study used secondary data to examine the association between depression and e-cigarette use among AIs who smoke. The findings contribute to a sparse literature, which, collectively, will be informative for future community-level preventative and treatment action for AI adults who smoke in a time when e-cigarette use is increasingly prolific among this health disparities population.

Data Availability

Restrictions apply to the availability of these data, which are owned by the Cherokee Nation, and are not publicly available. Data are, however, available with the explicit permission of the Cherokee Nation.