|Home | About | Journals | Submit | Contact Us | Français|
While sex differences in the nicotine withdrawal (NW) symptoms and craving (NC) have been extensively described in adult cigarette smokers, few studies have investigated these phenomena in adolescents. We investigated the effect of gender and hormonal contraception (HC) on NW and NC during the first 14 days of cessation in adolescent smokers using data from a randomized, placebo-controlled, double-blind trial of the transdermal nicotine replacement therapy for smoking cessation.
Analyses showed similar levels of NW severity in males and females, regardless of HC use. However, significantly higher NC was observed in females compared to males, (2.22±0.12 vs. 1.65±1.14; p=0.003). Further, females not using HC reported the highest level of NC (2.38±0.16) followed by females using HC (2.08±0.25) and males (1.71±0.16; p=0.007).
The current findings suggest that adolescent females experience similar NW severity to males, but have stronger NC. Further, the use of hormonal contraceptives may impact the severity of craving. Addressing these different symptoms in adolescents may be useful in increasing smoking cessation rates in this special population of smokers.
While gender differences in nicotine withdrawal (NW) and craving (NC) in adult cigarette smokers have been well studied (Leventhal, Waters, Boyd, Moolchan, Lerman, & Pickworth, 2007; Hogle & Curtin, 2006; Carpenter, Upadhyaya, LaRowe, Saladin, & Brady, 2006), sex differences in adolescent smokers have only recently been explored (Smith, Cavallo, Dahl, Wu, George & Krishnan-Sarin, 2008; Panday, Reddy, Ruiter, Berstrom & de Vries, 2007). Adolescent smokers quickly develop nicotine dependence and most experience NW during smoking cessation (DiFranza, Savageau, Fletcher et al, 2007). Female adolescent smokers may experience more severe NW than males (Panday et al, 2007), although withdrawal symptoms appear to resolve more quickly in females (Smith et al, 2008). As yet, the possible role of female sex hormones in nicotine withdrawal in adolescent smokers, including hormone contraceptives, has not been described. Benowitz and colleagues (2006) reported that women who used combined oral contraceptives or estrogen-only contraceptives had faster nicotine metabolism than those who took no hormones.
The animal literature supports a role of ovarian hormones in addictive drug behavior. There is strong evidence of drug self-administration via alterations in dopamine functioning in reward areas of the brain where estrogen has been suggested to enhance this dopamine reward system (Disshon & Dluzen, 1999; Thompson & Moss, 1994). However, clinical studies are not as clear about the role of ovarian hormones. Some retrospective trials (Perkins, Levine, Marcus et al, 2000; O’Hara, Portser & Anderson, 1989) have shown that women who happen to quit in the luteal phase have greater withdrawal. Yet other studies show no effect of cycle phase with intranasal nicotine (Marks, Pomerleau, & Pomerleau, 1999). Some retrospective analyses suggest that (Franklin, Ehrman, Lynch et al, 2008; Carpenter, Saladin, Leinbach, Larowe, & Upadhyaya, 2008) in women on nicotine replacement therapy (NRT) who quit in the luteal phase experience worse outcomes compared to follicular phase. On the other hand, another randomized trial of women not on NRT quitting in different phases of the menstrual cycle showed that women who quit in the follicular phase had poorer outcomes (Allen, Bade, Center, Finstad & Hatsukami, 2008). Although clinical results are mixed, there is a suggestion that ovarian hormones may play a role in women’s ability to quit.
Given this background, the purpose of the current report was to compare NW and NC in adolescent males, adolescent females not currently using hormonal contraceptives, and adolescent females using hormonal contraceptives (HC). The current report draws on data collected as part of a double-blind, placebo-controlled, randomized study of transdermal NRT in adolescent smokers (Hanson, Allen, Jensen & Hatsukami, 2003). We hypothesized that female adolescents would report more severe NW symptoms than male adolescents. Additionally, we hypothesized that adolescent females using hormonal contraceptives would exhibit more intense withdrawal symptoms compared to adolescent females not using hormonal contraceptives.
A complete description of study methods is provided elsewhere (Hanson et al, 2003). This study was reviewed and approved by the Institutional Review Board at the University of Minnesota. Participants under the age of 18 years were required to have parental written informed consent. Eligible participants: (1) were between the ages of 13–19; (2) smoked ≥10 cigarettes/day for at least 6 months; (3) were motivated to quit smoking (indicated by a score greater than or equal to 7, using a scale from 0 [not motivated at all] to 10 [very much motivated]; (4) did not use any other tobacco products more than once per week; and (5) were not currently using any form of nicotine replacement therapy (NRT). Exclusion criteria included: (1) current alcohol abuse or drug abuse problems; (2) severe emotional problems within the past year; (3) use of psychoactive medications (except those used to treat attention-deficit/hyperactivity disorder) within the past six months; and (4) were unable to use the nicotine patch due to medical contraindications.
Over the course of the 12-week study conducted from 1997 to 1999, participants attended a total of 13 visits at their respective school or at the Tobacco Use Research Center (TURC): visit -1 was orientation and a medical screening, visit 0 during week 1, when nicotine or placebo patch was randomly assigned, visits 1–2 during week 2 with visit 1 being the target quit date, visits 3–8 weekly for 6 weeks, and visits 9 and 10 biweekly for the final 4 weeks of treatment. Participants quit smoking the morning of visit 1. Data for this study was limited to the first two weeks post quit date when the greatest severity of withdrawal symptoms would be expected. At each treatment visit, both active patch and placebo groups received 10–15 minutes of individual cognitive behavioral therapy provided by a trained counselor. Expired-air carbon monoxide levels were measured at each visit using the Bedfont Micro Smokerlyzer device to determine abstinence. Participants were declared abstinent at a given time if (a) they attained CO levels ≤5 ppm at clinic visits and (b) they reported no cigarettes smoked on the daily diary. Salivary cotinine levels from participants were collected but only samples from participants in the placebo patch group were examined. This was used with the CO levels to determine false positives and negatives. A salivary cotinine level of ≤15ng/ml was considered abstinent (Hanson et al, 2003).
At the baseline visit, demographic and smoking history questionnaires and the Fagerstrom Test of Nicotine Dependence (FTND; Heatherton, Kozlowski, Frecker & Fagerstrom, 1991) were completed. Additionally, the participant’s medications and birth control use were documented in the participant’s medical record and verified by the study physician. The Minnesota Nicotine Withdrawal Scale (MNWS), a self-rated measure found to be a sensitive indicator of NW and NC, was the primary dependent variable in this study (Hughes & Hatsukami, 1986; Hughes & Hatsukami, 1998). NW was defined as the total of the 7 seven individual items, excluding craving, while NC was indicted by the separate craving item. Participants completed the MNWS at each clinic visit.
Data were analyzed using the Statistical Analysis System 9.1.3. (SAS Institute Inc., 2007). Values of p<0.05 were considered statistically significant, based on two-tailed tests. Male (n=43) and female smokers (n=57) were compared using t-tests and chi-square tests. Analyses focused on first 2 weeks of cessation when the greatest nicotine withdrawal severity was anticipated. The primary analytic strategy was repeated-measures analysis of covariance including as predictors: sex (or sex with females broken out by HC use), time, and their interaction. Covariates included patch condition, age, FTND, having previously quit for 24 hours or more, and average values of the MNWS during the time preceding the analysis period (visit -1 and visit 0). Smoking status was treated as a time varying covariate for days 1, 7, and 14.
Comparison of male and female participants revealed no significant differences on demographic and smoking history variables (Table 1).
In the first 14 days of treatment, females, on average (± standard error) reported similar levels of total withdrawal as seen in males (7.7±0.60 vs. 6.4±0.66, respectively; F(1,76) = 2.30, p = 0.1333. Analysis of NC revealed a sex effect on craving such that females reported higher craving than males (2.22± 0.12 vs. 1.65±1.14, respectively; F(1,76.7)=9.15, p = 0.0034. NC tended to decline in the time since the quit day, F(2, 73.3) = 9.10, p = 0.0003. See Figure 1.
Of the 57 female participants, 18 (32%) reported using hormonal contraception, 12 (67%) combined, 6 (33%) progesterone-only. The above analyses were repeated, partitioning sex into male, female without HC, and female with HC. Results showed no effect of hormonal contraceptives on total withdrawal or withdrawal excluding craving. However, there was evidence that hormonal contraceptives affected craving as a single item such that females with HC reported the highest level of craving (2.38±0.16), followed by females without HC (2.11± 0.25) then males (1.71± 0.16; F(2, 75.3) = 5.20, p = 0.0077). Again, NC tended to decline in the time since the quit day, F(2, 76.2) = 7.33, p = 0.0012. See Figure 2.
At variance with recent reports, adolescent females in our sample reported similar NW to adolescent males (Smith et al, 2008; Panday et al, 2007). However, females reported higher levels of craving than males. Studies have shown that nicotine replacement is not as effective for women compared to men (Hatsukami, Skoog, Allen & Bliss, 1995; Gourlay, Forbes, Marriner, Pethica & McNeil, 1994; Bjornson, Rand, Connett, et al, 1995; Swan, Jack & Ward, 1997; Perkins, 1999) and this may have partially contributed to the observed increased craving seen in the female adolescents. Furthermore studies show (Perkins, 1999) that nicotine dependence in women may relate more to sensory and social cues.
Withdrawal severity was similar in female adolescents, regardless of hormonal contraceptive use. However, craving as a separate item varied according to hormone use, such that those using hormonal contraceptives experienced greater craving than those who were not using hormonal contraceptives. Perhaps hormonal contraceptives effect on craving, not withdrawal, is more pronounced. There is an overlap between premenstrual symptoms and withdrawal symptoms (Allen, Hatsukami, Christianson, & Nelson, 1999) and since hormonal contraceptives can diminish premenstrual symptoms, withdrawal symptoms may be blunted such that a difference between adolescents with and without hormonal contraceptives would not be discernable. Yet craving acts more as an independent item and does not overlap with premenstrual symptoms and therefore might be more prominent.
The observed effects of hormonal birth control on craving may be limited by a variety of factors other than the underlying hormonal mechanisms of birth control use. First, only a small number of females were using hormonal contraceptives while enrolled in the study. Because this was an observational study, females were not randomized into groups of hormonal contraceptive users versus non-users. Given these small numbers we were unable to assess differences between those using combined estrogen/progesterone hormonal contraceptives compared to those using progesterone-only hormonal contraceptives. Second, the hormonal contraceptives that were utilized during the study were comprised of varying doses of different hormones. For the purposes of this study, a female’s use of birth control was classified simply as a “hormonal contraceptive use” or not. Further, some of the combined contraceptives were of varying chemical compositions throughout the menstrual cycle (triphasic oral contraceptives) while others had constant chemical compositions throughout the menstrual cycle (monophasic oral contraceptives). Future studies using a consistent monophasic birth control pill to deliver a constant dose of estrogen and progesterone would provide a stronger model to measure if ovarian hormones have an effect on withdrawal.
This report has some additional limitations. The numbers of females on HC and the variation of the type of HC limited the power of the groups. The retrospective nature limits interpretability of the effect of HC.
This study provides new information on gender differences in withdrawal and craving in adolescents attempting to quit smoking using nicotine replacement. The data suggests that female adolescents experience similar NW as males but have higher levels of craving. The use of HC may also impact the level of craving in females although this needs to be replicated in a larger study controlled for HC use. However it is prudent to address these different symptoms in adolescent males and females to increase smoking cessation outcomes in adolescents
This research was supported by NIDA grants R01-DA-014538, P50-DA-09259, and P50-DA-13333. The first author was supported by grants from the Minnesota Medical Foundation and the Minnesota Academy of Family Physicians. Dr. Mooney is supported by a NIDA Career Development award K01-DA-019446.
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.