Abstract
The objective of the current study was to examine demographic, tobacco-related, and psychosocial factors associated with cotinine concentration in a group of 256 male smokeless tobacco (ST) users living in the Ohio Appalachian region. Participants completed a survey that included questions on: 1) current and past tobacco use behaviors; 2) demographics; 3) tobacco dependence; 4) decisional balance; 5) health behaviors; and 6) perceived stress and depressive symptoms. Saliva samples were obtained for measurement of cotinine. The variables related to salivary cotinine concentration in the multiple regression model were age, marital status, occupation, quit attempts, years of ST use, and tobacco dependence score (adjusted R² = 0.24). Among the 199 snuff only users, cotinine concentration was positively related to age, being divorced/widowed/separated, no quit attempts in the previous year, dependence score, and brand of snuff (adjusted R² = 0.29). This is one of the largest studies to examine influences beyond topography on cotinine concentration in a group of rural ST users. These findings suggest that smokeless tobacco users and smokers share some similarities with respect to tobacco dependence.
Keywords: Cotinine, smokeless tobacco, dependence
Introduction
Cotinine is the major metabolite of nicotine and has become the standard marker of nicotine exposure (Jarvis et al., 1987). However, data also indicate that it is a marker of nicotine dependence. The Society for Research on Nicotine and Tobacco (SRNT) Subcommittee on Biochemical Verification published guidelines for use of biomarkers in tobacco control studies (SRNT, 2002), and one of the recommendations is to use cotinine as an indicator of addiction severity. However, it should be noted that the basis for this recommendation are results obtained from the literature on smoking, not smokeless tobacco use. Limited information exists on the correlates of cotinine among rural male smokeless tobacco (ST) users, a group with a high prevalence of use. Among ST users, cotinine concentration has been positively associated with age, years of ST use, dips per day, duration of a dip, and total dipping time (Lemmonds et al., 2005; Hatsukami et al., 1988). In general, most of the studies that have examined factors associated with cotinine concentration among ST users have been small and primarily focused on characterizing topography using diaries (Lemmonds et al., 2005; Hatsukami et al., 1988). As a logical extension of these studies, a cross-sectional field study that allows for an estimation of the usual concentration of cotinine and examination of additional factors is warranted. Demographic and psychosocial factors associated with cotinine concentration among ST users have not been reported in the literature. Knowledge of such relationships is important when developing effective cessation interventions (Fiore et al., 2000).
The objectives of the analyses presented in this paper were to describe demographic, tobacco-related, and psychosocial factors associated with baseline cotinine concentration in a group of 256 men enrolled in a ST cessation study. This study adds to the literature on the correlates of cotinine and dependence among ST users by reporting data that were collected from a large group of rural Appalachian residents using field study methodology.
Methods
Setting
This study was conducted in two Ohio Appalachia counties: Ross County and Muskingum County. The Ohio Appalachian population is medically underserved and experiences high rates of unemployment, poverty, educational disparities, and poor health (Friedell et al., 1998). Poverty is estimated at 13.5% (10.6% in Ohio) and 22% of the residents have not attained a high school diploma (17% in Ohio). With respect to health insurance, 14.6% of Appalachian residents are estimated as uninsured (11.7% in Ohio) and, among those insured, 65.5% do not have adequate coverage (59.7% in Ohio) (Dorsky et al., 2002). Cancer mortality rates are high, particularly for tobacco-related cancers (MMWR, 2002). Smokeless tobacco use is estimated at 9.5% in Ohio Appalachia (3.6% in Ohio) (Renaud et al., 2006) and there is a long history of tobacco farming in the entire Appalachian region. Approximately 97% of all burley tobacco is grown in Appalachia (Wood, 1998) and in Ohio tobacco farming occurs in 16 of the Appalachian counties. Consequently, the social norms surrounding tobacco use are influenced by the reliance on tobacco for economic well-being (Denham and Rathbun, 2005).
Subjects
The study protocol and research procedures were approved by The Ohio State University Institutional Review Board. The subjects included in this analysis were enrolled in a smokeless tobacco cessation study that involved an appointment with a dentist at a rural health department clinic. After receiving a brief oral examination and strong advice from the dentist to quit using ST, the participant was assigned to either lay-led counseling with nicotine replacement therapy or usual-care control. All men were recruited through ad placement at community agencies in the two Appalachian Ohio counties. Eligible participants were male county residents, aged 18 and older, who were daily users of ST with no contraindications for nicotine replacement therapy (NRT) use.
Measures
The baseline questionnaire consisted of the following sections: 1) current and past tobacco use behaviors; 2) demographics; 3) dental health; 4) tobacco dependence; 5) decisional balance; 6) health behaviors; and 7) perceived stress and depressive symptoms. The tobacco use section captured data about use of various forms of tobacco, including snuff, chewing tobacco, and cigarettes. Ever use was measured, as well as details about current use such as dips per day, duration of each dip, time between dips, and type of ST used most often. Tobacco dependence was measured with a modification of the Fagerström Test of Nicotine Dependence (FTND) developed by Severson & Gordon (1997), called the Smokeless Tobacco Dependence Scale. The range of this scale is 2–9, with a higher number implying greater tobacco dependence. The psychometric properties of the ST dependence scale have been examined and reported (Ferketich et al., 2007). Briefly, the correlation between the total score and salivary cotinine concentration was moderate among the ST only users (r = 0.34), whereas it was lower (r = 0.19) among the ST + cigarette users. Among ST only users, the coefficient alpha for the scale was 0.40; however, it was higher among the ST + cigarettes group (alpha = 0.61). Decisional balance was measured using a modification of the smoking decisional balance scale (Velicer et al., 1985). Briefly, this scale consists of two 10-item subscales: a pro subscale evaluating positive aspects of tobacco use and a con subscale evaluating negative aspects. Decisional balance is calculated as the pro score minus the con score; thus higher scores indicate more favorable views of tobacco use. Depressive symptoms were measured with the Center for Epidemiologic Studies Depression (CES-D) scale, a 20-item questionnaire (Radlor, 1977). Stress was assessed with the 14-item Perceived Stress Scale (Cohen et al., 1982). For both of these scales, higher scores indicate more symptoms (i.e. more symptoms of depression and higher levels of perceived stress, respectively).
A saliva sample was also obtained from each man at baseline. Saliva was collected and was processed in the county health department, refrigerated at 4° C, and brought weekly to our laboratory for storage at −85°C until analyses were conducted. Cotinine was extracted from saliva using a gas chromatography/mass spectrometry (GC-MS) technique (Hariharan and VanNoord, 1991).
Statistical analysis
The objective of the analysis was to describe the factors associated with cotinine concentration in ST users. The first part of the analysis involved calculating descriptive statistics, including the mean, standard deviation, median, and range of cotinine concentration, by participant characteristics. For continuous measures, such as scores on the scales, scatterplots were examined and a correlation coefficient was estimated to determine the linear relationship between the score and cotinine concentration. If a linear relation was not apparent, then the variable was categorized. Additionally, if a variable was not truly continuous in nature in the sense that a one-unit increase has the same meaning over the range of possible values (e.g. quit attempts), then an ordinal variable was created. A linear regression model was fit to the data to determine which factors were significantly related to salivary cotinine among all ST users. A similar analysis was performed after limiting the data set to snuff only users. The method of forward selection was used for inclusion of variables and the significance level was set to 0.05; variables were not removed if the p-value increased following inclusion of other covariates. The two model assumptions, normality and equal variances of the residuals, were examined. Transformations, such as the natural log or square root transformation, were applied to the cotinine data when necessary to meet the assumptions.
Results
A total of 256 men completed the baseline survey. The average age of the sample was 34 years and the median cotinine concentration was 460 ng/mL (range 17–2469 ng/mL). Table 1 contains the frequency distributions for the demographic characteristics of the sample. Overall, it was a group with a low socioeconomic status: 14.5% had not received a high school diploma, 28.1% were unemployed, 28.9% had an income under $25,000, and only 63.7% had private health insurance. The majority of participants (86.7%) spent the first 18 years of their life in the Appalachian region. Simple linear regression models were fit to the data to estimate the association between each demographic variable and cotinine concentration. The model assumptions were not met; therefore the results are presented for the square-root transformed cotinine data, since this transformation resulted in residuals that more closely met the model assumptions. Being divorced or widowed or separated, working full time, and being a skilled laborer were associated with significantly higher salivary cotinine concentration.
Table 1.
Demographic Characteristics of 256 Males in a Smokeless Tobacco Cessation Study
| Variable | Number (%) in Sample | Mean ± SD Cotinine (ng/mL) | Median Cotinine (ng/mL) | Cotinine Range (ng/mL) |
|---|---|---|---|---|
| Educational Attainment | ||||
| < HS diploma | 37 (14.5%) | 483 ± 293 | 428 | 33–1264 |
| HS diploma | 96 (37.5%) | 584 ± 414 | 485 | 17–2469 |
| > HS diploma | 123 (48.0%) | 532 ± 341 | 440 | 29–2077 |
| Marital Status*** | ||||
| Divorced/Widow/Separated | 46 (18.0%) | 682 ± 520 | 544 | 17–1596 |
| Married/Living with partner | 161 (62.9%) | 562 ± 319 | 479 | 73–2469 |
| Never Married | 49 (19.1%) | 364 ± 254 | 277 | 33–1028 |
| Employment Status* | ||||
| Full-time worker | 173 (67.5%) | 581 ± 364 | 478 | 17–2097 |
| Part-time worker | 11 (4.3%) | 389 ± 264 | 284 | 118–937 |
| Unemployed | 72 (28.1%) | 482 ± 368 | 387 | 33–2469 |
| Occupation** | ||||
| Professional | 40 (15.6%) | 505 ± 322 | 434 | 29–1179 |
| Skilled labor | 128 (50.0%) | 615 ± 361 | 524 | 53–2097 |
| Unskilled labor | 88 (34.4%) | 460 ± 371 | 392 | 17–2469 |
| Income | ||||
| ≤ $25,000 | 72 (28.9%) | 501 ± 403 | 443 | 38–2469 |
| $25,001–50,000 | 95 (38.2%) | 550 ± 344 | 460 | 21–2077 |
| > $50,000 | 82 (32.9%) | 590 ± 357 | 504 | 17–1596 |
| Insurance Type | ||||
| Medicare | 10 (3.9%) | 632 ± 342 | 577 | 213–1264 |
| Medicaid | 34 (13.3%) | 444 ± 296 | 409 | 33–1037 |
| Private insurance | 163 (63.7%) | 566 ± 351 | 476 | 17–2077 |
| Self-pay | 49 (19.1%) | 529 ± 446 | 394 | 92–2469 |
| County Residence thru age 18 | ||||
| Appalachian | 222 (86.7%) | 540 ± 371 | 460 | 17–2469 |
| Non-Appalachian | 34 (13.3%) | 578 ± 327 | 461 | 38–1193 |
p < 0.05 obtained from the nonparametric Wilcoxon Rank Sum test
p < 0.01 obtained from the nonparametric Wilcoxon Rank Sum test
p < 0.0001 obtained from the nonparametric Wilcoxon Rank Sum test
With respect to tobacco use characteristics (Table 2), 24.2% of the men were under age 10 when they started using ST regularly, 31% reported using ST for more than 20 years, 15.6% smoked in addition to daily ST, 87.9% used snuff only, and 57.1% had made at least one quit attempt in the past year. A small percentage of men reported leaving ST in over night (8.4%) and almost half (45.3%) reported that the time between consecutive dips was 30 minutes or less. Among the snuff users, the three most popular brands were Copenhagen (40%), Skoal (28.4%), and Timberwolf (15.8%). On average, snuff users consumed 4.6 cans per week and each dip remained in the mouth for 70 minutes. Cotinine concentration was positively related to years of ST use, no quit attempts in the past year, a shorter time between dips, and the tobacco dependence score. Additionally, among snuff only users, mean cotinine concentration was higher among men who used Red Seal and Copenhagen.
Table 2.
Tobacco Use Characteristics of 256 Males in a Smokeless Tobacco Cessation Study
| Variable | Number (%) in Sample | Mean ± SD Cotinine (ng/mL) | Median Cotinine (ng/mL) | Cotinine Range (ng/mL) |
|---|---|---|---|---|
| Age at initiation | ||||
| ≤ 10 years | 62 (24.2%) | 610 ± 387 | 491 | 33–2077 |
| 11–17 years | 142 (55.5%) | 516 ± 336 | 445 | 17–2097 |
| 18 or older | 52 (20.3%) | 546 ± 409 | 454 | 29–2469 |
| Years of ST use*** | ||||
| < 10 | 74 (29.0%) | 409 ± 323 | 331 | 29–2097 |
| 10–19 | 102 (40.0%) | 524 ± 232 | 466 | 17–1484 |
| ≥ 20 | 79 (31.0%) | 697 ± 438 | 587 | 39–2469 |
| Other tobacco use | ||||
| ST only | 216 (84.4%) | 560 ± 370 | 463 | 33–2097 |
| ST + cigarettes | 40 (15.6%) | 460 ± 332 | 438 | 17–2469 |
| Type of ST | ||||
| Snuff only | 225 (87.9%) | 542 ± 353 | 463 | 21–2469 |
| Chew only | 23 (9.0%) | 578 ± 478 | 449 | 17–2077 |
| Snuff + chew | 8 (3.1%) | 520 ± 362 | 397 | 211–1352 |
| Quit attempts in past year*** | ||||
| 0 | 105 (42.9%) | 672 ± 368 | 622 | 29–2077 |
| ≥ 1 | 140 (57.1%) | 457 ± 345 | 395 | 17–2469 |
| Time between dips** | ||||
| ≤ 30 minutes | 116 (45.3%) | 620 ± 400 | 501 | 21–2077 |
| > 30 minutes | 140 (54.7%) | 483 ± 321 | 417 | 17–2469 |
| Leave dip in over night | ||||
| Yes | 20 (8.4) | 663 ± 530 | 627 | 38–2469 |
| No | 217 (91.6) | 530 ± 328 | 459 | 21–2097 |
| Snuff Only Users (n=190)** | ||||
| Type of snuff | ||||
| Copenhagen | 76 (40.0%) | 644 ± 381 | 618 | 78–2469 |
| Skoal | 54 (28.4%) | 464 ± 339 | 343 | 21–1205 |
| Timberwolf | 30 (15.8%) | 435 ± 241 | 422 | 76–937 |
| Red Seal | 15 (7.9%) | 675 ± 273 | 676 | 262–1082 |
| Other | 15 (7.9%) | 570 ± 377 | 443 | 154–1484 |
p < 0.01 obtained from the nonparametric Wilcoxon Rank Sum test
p < 0.0001 obtained from the nonparametric Wilcoxon Rank Sum test
Results concerning the continuous variables (age, psychosocial scale variables, and tobacco dependence score) are listed in Table 3. Age and the tobacco dependence score were significantly and positively correlated with cotinine concentration. Additionally, all three psychosocial variables were significantly associated with cotinine concentration; however, the correlation coefficients were small in magnitude. The large sample size likely contributed to the significant result.
Table 3.
Descriptive Statistics for Continuous Variables among 256 Males in a Smokeless Tobacco Cessation Study
| Variable | Mean ± SD | Correlation with Cotinine |
|---|---|---|
| Age*** | ||
| Center for Epidemiologic Studies Depression Scale* | 34.1 ± 11.1 | 0.36 |
| 9.9 ± 7.8 | −0.14 | |
| Perceived Stress Scale* | 20.6 ± 8.3 | −0.15 |
| Decisional Balance (Cons of tobacco – Pros of tobacco)* | 0.3 ± 8.7 | −0.13 |
| Tobacco Dependence Score*** | 5.3 ± 1.5 | 0.31 |
| Snuff Only Users (n=190) | ||
| Snuff consumption | 4.6 ± 3.3 cans per week | 0.10 |
| Time each dip is in mouth | 70 ± 67 minutes | 0.12 |
p < 0.05 from the Spearman Rank Correlation coefficient test
p < 0.001 from the Spearman Rank Correlation coefficient test
The multiple regression model results are presented in Table 4. The variables that were included in the overall final model were age, marital status, occupation, quit attempts, tobacco dependence score, and CES-D score. Cotinine concentration was positively related to being divorced/widowed/separated (compared to being single or married), having a skilled labor occupation (compared to professionals), having no quit attempts in the previous year, having a higher tobacco dependence score, and having a lower CES-D score. The adjusted R² from the model was 0.27 and the diagnostics performed did not suggest problems with outliers or influential observations. Among the 199 snuff only users, cotinine concentration was positively related to age, no quit attempts, and tobacco dependence score, and negatively related to CES-D score (Table 5). Brand of snuff was also related to cotinine concentration in this model. The adjusted R² was 0.36 and diagnostics indicated a good fit.
Table 4.
Final Multiple Regression Model for Square-Root Transformed Cotinine in 256 Male Smokeless Tobacco Users
| Variable | Parameter Estimate | S.E. | t-statistic | p-value |
|---|---|---|---|---|
| Demographic Characteristics | ||||
| Age | 0.17 | 0.04 | 3.91 | 0.0001 |
| Marital Status | ||||
| Divorced/Widowed/Separated | -- | |||
| Married | −2.36 | 1.19 | −1.98 | 0.049 |
| Single | −3.42 | 1.54 | −2.22 | 0.028 |
| Occupation | ||||
| Professional | -- | |||
| Skilled labor | 2.92 | 1.24 | 2.35 | 0.020 |
| Unskilled labor | 1.63 | 1.38 | 1.19 | 0.236 |
| Tobacco Use Characteristics | ||||
| Quit attempts in the past year | ||||
| 0 | -- | |||
| ≥ 1 | −3.27 | 0.91 | −3.59 | 0.0004 |
| Tobacco Dependence Score | 1.27 | 0.29 | 4.41 | <0.0001 |
| Psychosocial Characteristic | ||||
| Center for Epidemiologic Studies Depression Scale | −0.14 | 0.06 | −2.42 | 0.016 |
Adjusted R² = 0.27
Table 5.
Final Multiple Regression Model for Square-Root Transformed Cotinine in 199 Male Snuff Tobacco Users
| Variable | Parameter Estimate | S.E. | t-statistic | p-value |
|---|---|---|---|---|
| Demographic Characteristic | ||||
| Age | 0.32 | 0.05 | 6.16 | <0.0001 |
| Tobacco Use Characteristics | ||||
| Quit attempts in the past year | ||||
| 0 | -- | |||
| ≥ 1 | −2.13 | 0.95 | −2.23 | 0.027 |
| Tobacco Dependence Score | 1.44 | 0.32 | 4.46 | <0.0001 |
| Type of Snuff | ||||
| Red Seal | -- | |||
| Copenhagen | −2.04 | 1.73 | −1.18 | 0.24 |
| Skoal | −4.60 | 1.77 | −2.60 | 0.010 |
| Timberwolf | −4.21 | 1.91 | −2.20 | 0.029 |
| Other | −4.64 | 2.23 | −2.07 | 0.039 |
| Psychosocial Characteristic | ||||
| Center for Epidemiologic Studies Depression Scale | −0.17 | 0.06 | −2.85 | 0.005 |
Adjusted R² = 0.36
Discussion
To our knowledge, this is one of the largest studies to examine influences beyond tobacco use characteristics on cotinine concentration in a group of ST users. Another unique feature of this study is the focus on rural ST users. While several demographic, social, and psychosocial variables were related to cotinine concentration in the univariate analyses, only age, occupation, marital status, and CES-D score were retained in the final model that included all ST users. With respect to occupation, men who worked as skilled laborers had a significantly higher cotinine concentration compared to men in professional occupations. Occupation has not been previously associated with smokeless tobacco use; however, it has been related to smoking in several studies, with laborers or blue-collar occupations smoking at higher rates compared to professionals (Shavers et al., 2005). While employment status was not retained in the final model, it was significant in the univariate model with a higher cotinine concentration among employed men. It is possible that men who work full-time are using more ST because smoking is prohibited at the worksite. Recent data from the Ohio Adult Tobacco Survey suggest that 69% of Appalachian residents report having official policies that ban smoking in public/common areas at their worksite and 75% report having policies that completely ban smoking in work areas (Renaud et al., 2006). Marital status was also retained in the final model and the results indicated that men who were single or married had a significantly lower cotinine concentration compared to men who were divorced, widowed, or separated. From studies reported in the smoking literature it appears as though marital disruption is related to tobacco consumption and an inability to quit (Nystedt, 2006), and being married is associated with greater success at cessation among men (Broms et al., 2004).
Years of tobacco use, quit attempts in the previous year, and time between dips were significantly related to cotinine concentration in the univariate models; however, only quit attempts remained in the final model. Hatsukami and colleagues reported that the number of dips per day, the average duration of a dip, the total duration of ST use in a day, and the mean inter-dip interval were significantly related to cotinine concentration in a group of 56 male Copenhagen users (Hatsukami et al., 1988). In a later study, only average total daily dip duration was related to cotinine in a group of 54 male ST users (Lemmonds et al., 2005).
Similar associations were found among snuff only users, although occupation and marital status were not retained in the final model and snuff brand was significantly associated with cotinine concentration. Red Seal® and Copenhagen® users had the highest concentration of cotinine, followed by the other brands. The men who used these two brands were longer-term ST users in general, compared to the men who used the other brands (81.3% versus 66.7% had used ST for more than 10 years). This result supports the work of Tomar and colleagues on brand switching (Tomar et al., 1995). They found that over time youth in their study switched from low nicotine snuff to high nicotine brands, specifically Copenhagen. In the current study, the association between snuff brand and cotinine concentration remained significant even after controlling for years of ST use.
In both models, the tobacco dependence scale score was associated with cotinine, which indicates that these two measures of dependency are significantly related among ST users. These results are consistent with the findings reported in the smoking literature (Heatherton et al., 1991). Decisional balance, however, was not strongly associated with cotinine concentration in the univariate model. While the direction of the correlation coefficient suggested that men who associated ST use with more cons than pros had lower cotinine concentrations, the magnitude of the coefficient was very small and it did not remain significant during model building.
Interestingly, the associations between cotinine and both depressive symptoms and stress were not strong in the univariate models; moreover, the correlation coefficients were negative. The CES-D score was significantly related to cotinine concentration in the final model; however, as before, the parameter estimate was negative which suggests higher cotinine concentrations among men with lower CES-D scores (i.e. fewer symptoms of depression). These findings for depressive symptoms and perceived stress are contrary to those from the smoking literature, where both variables are associated with higher levels of tobacco dependence (John et al., 2006; Hu et al., 2006). Thus, in this cohort of male ST users, it does not appear as though more tobacco is used among men who report higher levels of perceived stress and/or depressive symptoms. A possible explanation for our finding of no relation between cotinine concentration and both perceived stress and depressive symptoms is that the rewards are different for ST users compared to cigarette smokers. One reward, relaxation, is reported by approximately 75% of smokers; however, only 50% of ST users state that feeling calm or relaxed is a reason for using ST (CDC, 1994; Hatsukami and Severson, 1999).
The main limitation of this study relates to the sample of smokeless tobacco users. All of the users were male volunteers who were interested in joining a dental clinic tobacco cessation study and they were all from the Appalachian region in one state, which could limit the generalizability of the findings. A second limitation arises because there is a question of whether cotinine is a valid marker of dependence among ST users. The behavior of swallowing tobacco juice, which was reported by approximately 34% of the men in this study, may artificially raise the concentration of cotinine among individuals who swallow the tobacco juices. Ebbert and colleagues reported that tobacco juice swallowing was related to a higher concentration of serum cotinine, but not to a higher concentration of serum nicotine (Ebbert et al., 2004). Thus, there may be some men who had a high cotinine concentration because of engaging in this behavior and not because they are more dependent on ST. The tobacco dependence questionnaire includes a question on swallowing tobacco juice; therefore this behavior was controlled for in the analysis as part of the total tobacco dependence score.
In conclusion, we noted several factors related to cotinine concentration among ST users. Some factors are similar to those reported in the cigarette smoking literature, including occupation, marital status, and tobacco dependence scale score. Depression and perceived stress were negatively associated with cotinine, and only the former was retained in the final model. This association deserves further examination as it is not consistent with studies published in the smoking literature. It is possible that ST use may operate differently with regard to affective states.
Acknowledgement
This study was supported by a grant from the National Institutes of Health (R01 DE13926). The contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Broms U, Silventoinen K, Lahelma E, Koskenvuo M, Kaprio J. Smoking cessation by socioeconomic status and marital status: The contribution of smoking behavior and family background. Nicotine & Tobacco Research. 2004;6:447–455. doi: 10.1080/14622200410001696637. [DOI] [PubMed] [Google Scholar]
- Centers for Disease Control. Reasons for tobacco use and symptoms of nicotine withdrawal among adolescent and young adult tobacco users – United States, 1993. Morbidity and Mortality Weekly Report. 1994;43:745–750. [PubMed] [Google Scholar]
- Cohen S, Kamarch T, Mermelstein R. A global measure of perceived stress. Journal of Health and Social Behavior. 1982;24:385–396. [PubMed] [Google Scholar]
- Denham SA, Rathbun A. Appalachian Rural Health Institute; Community health assessment Appalachia: an overview of health concerns and health literacy. 2005 January;
- Dorsky D, Ramsini W, Holtzhauer F. Ohio Department of Health; Access to Health Care for Ohio Adults, 1998. 2002
- Ebbert JO, Dale LC, Nirelli LM, Schroeder DR, Moyer TP, Hurt RD. Cotinine as a biomarker of systemic nicotine exposure in spit tobacco users. Addictive Behaviors. 2004;29:349–355. doi: 10.1016/j.addbeh.2003.08.046. [DOI] [PubMed] [Google Scholar]
- Ferketich AK, Wee AG, Shultz J, Wewers ME. A Measure of Nicotine Dependence for Smokeless Tobacco Users. Addictive Behaviors. 2007 Jan 11; doi: 10.1016/j.addbeh.2007.01.005. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fiore MC, Bailey WC, Cohen SJ, et al. Treating Tobacco Use and Dependence. Clinical Practice Guideline. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service; 2000. Jun, [Google Scholar]
- Friedell GH, Linville LH, Hullet S. Cancer control in rural Appalachia. Cancer. 1998;83:1868–1871. [Google Scholar]
- Hariharan M, VanNoord T. Liquid-chromatographic determination of nicotine and cotinine in urine of passive smokers: Comparison with gas chromatography with a nitrogen-specific detector. Clinical Chemistry. 1991;37:1276–1280. [PubMed] [Google Scholar]
- Hatsukami DK, Keenan RM, Anton DJ. Topographical features of smokeless tobacco use. Psychopharmacology. 1988;96:428–429. doi: 10.1007/BF00216076. [DOI] [PubMed] [Google Scholar]
- Hatsukami DK, Severson HH. Oral spit tobacco: addiction, prevention and treatment. Nicotine and Tobacco Research. 1999;1:21–44. doi: 10.1080/14622299050011131. [DOI] [PubMed] [Google Scholar]
- Heatherton T, Kozlowski L, Frecker R, Fagerström K. The Fagerström test for nicotine dependence: a revision of the Fagerström tolerance questionnaire. British Journal of Addiction. 1991;86:1119–1127. doi: 10.1111/j.1360-0443.1991.tb01879.x. [DOI] [PubMed] [Google Scholar]
- Hu MC, Davies M, Kandel DB. Epidemiology and correlates of daily smoking and nicotine dependence among young adults in the United States. Am J Public Health. 2006;96:299–308. doi: 10.2105/AJPH.2004.057232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jarvis MJ, Russell MAH, Saloojee Y. Expired air carbon monoxide: a simple test of tobacco smoke intake. British Med J. 1989;281:484–485. doi: 10.1136/bmj.281.6238.484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- John U, Riedel J, Rumpf HJ, Hapke U, Meyer C. Associations of perceived work strain with nicotine dependence in a community sample. Occup Environ Med. 2006;63:207–211. doi: 10.1136/oem.2005.020966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lemmonds CA, Hecht SS, Jensen JA, Murphy SE, Carmella SG, Zhang Y, et al. Smokeless tobacco topography and toxin exposure. Nicotine and Tobacco Research. 2005;7:469–474. doi: 10.1080/14622200500135640. [DOI] [PubMed] [Google Scholar]
- MMWR. MMWR; Cancer death rates-Appalachia, 1994–1998. 2002;51:527–529. [PubMed]
- Nystedt P. Marital life course events and smoking behaviour in Sweden 1980–2000. Social Science in Medicine. 2006;62:1427–1442. doi: 10.1016/j.socscimed.2005.08.009. [DOI] [PubMed] [Google Scholar]
- Radlor LS. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
- Renaud J, Ray S, Homsi G, Salib S, Hersey J. RTI International; Results of the 2005 Ohio Adult Tobacco Survey: Regional Report. 2006 June;
- Severson HH, Gordon JS. Enough Snuff: A Guide to Quitting Smokeless Tobacco. 4th ed. Eugene, OR: Applied Behavior Science Press; 1997. [Google Scholar]
- Shavers VL, Lawrence D, Fagan P, Gibson JT. Racial/ethnic variation in cigarette smoking among the civilian US population by occupation and industry, TUS-CPS 1998–1999. Preventive Medicine. 2005;41:597–606. doi: 10.1016/j.ypmed.2004.12.004. [DOI] [PubMed] [Google Scholar]
- SRNT Subcommittee on Biochemical Verification. Biochemical verification of tobacco use and cessation. Nicotine & Tobacco Research. 2002;4:149–159. doi: 10.1080/14622200210123581. [DOI] [PubMed] [Google Scholar]
- Velicer W, DiClemente C, Prochaska J, Brandenburg N. Decisional balance measure for assessing and predictiong smoking status. Journal of Personality and Social Psychology. 1985;48:1279–1289. doi: 10.1037//0022-3514.48.5.1279. [DOI] [PubMed] [Google Scholar]
- Wood LE. Appalachian Regional Commission; The Economic Impact of Tobacco Production in Appalachia. 1998 November;