Smokers' Narrative Accounts of Quit Attempts: Aids

0 downloads 0 Views 54KB Size Report
the actual change process. An understanding of ... narrative accounts should be used to supplement traditional data collection ... studied using full accounts of people's lives (Kaufman, 1986;. McAdams .... nominal variables (e.g., gender, marital status). We used a ..... Community narratives and personal stories in. Alcoholics ...
Psychology of Addictive Behaviors 2006, Vol. 20, No. 2, 219 –224

Copyright 2006 by the American Psychological Association 0893-164X/06/$12.00 DOI: 10.1037/0893-164X.20.2.219

Smokers’ Narrative Accounts of Quit Attempts: Aids and Impediments to Success Todd M. Helvig, Linda Carter Sobell, Mark B. Sobell, and Edward R. Simco Nova Southeastern University In this study, the authors used cigarette smokers’ narratives describing their quit attempts to understand factors related to the change process. Maintained quitters (MQs, n ⫽ 59) and temporary quitters (TQs, n ⫽ 47) wrote autobiographical narratives describing their most serious (TQs) or last (MQs) quit attempt. Two types of content analysis were used to analyze the reports: (a) dichotomous ratings of the presence or absence of an event and (b) computerized content analysis of event or word frequency. The valence (anti– or pro–smoking cessation) of change factors was also examined. MQs wrote significantly more affective statements than did TQs. When valence was examined, MQs made significantly more pro– smoking cessation social support, cognitive, and affective statements than TQs did, and TQs made significantly more anti–smoking cessation social support and affective statements than MQs did. Keywords: cigarettes, smoking, tobacco cessation, mechanisms of change

narrative accounts should be used to supplement traditional data collection approaches (Baumeister et al., 1990; Gergen & Gergen, 1998; Klingemann, 1991, 1992). Although behavior change can be studied using full accounts of people’s lives (Kaufman, 1986; McAdams, 1985; Rosenberg, 1988), a more manageable and less laborious strategy involves the use of micronarratives, where respondents write a short, focused statement about a specific behavior change (e.g., gambling, drug abuse) or a major life event (e.g., marriage, childbirth). In social psychology, micronarratives have been used to explore subjective perceptions of many behaviors and phenomena (e.g., anger, masochism, failed relationships, love; Baumeister et al., 1990; Baumeister & Stillwell, 1992; McAdams, 1994). Addiction researchers have also used micronarratives to better understand processes of behavior change and relapse (Bottorff, Johnson, Irwin, & Ratner, 2000; Burman, 1997; Ellingstad, Sobell, Sobell, Eickleberry, & Golden, 2006; Hanninen & Koski-Ja¨nnes, 1999; Heatherton & Nichols, 1994; Humphreys, 2000; Klingemann, 1991; McIntosh & McKeganey, 2000; Parry, Fowkes, & Thomson, 2001; Sobell et al., 2001). Narrative accounts allow people to provide an understanding of how they perceive their changes to have occurred (Baumeister, 1996; Harvey, Weber, & Orbuch, 1990; McAdams, 1994). Because such accounts retrospectively gather data, they have a potential for bias (e.g., misattributions, distortions, self-aggrandizing, responses to demand characteristics). Although prospective studies would avoid such bias, they have the disadvantage of requiring a major commitment of resources (i.e., a large sample is needed because only some people will demonstrate change) and time (i.e., longitudinal studies can extend over years). Clearly, both prospective and retrospective studies have a place in advancing psychologists’ understanding of behavior change processes. In the present exploratory study, we, using a methodology derived from that reported by Heatherton and Nichols (1994), gathered micronarratives retrospectively from individuals who either successfully or unsuccessfully stopped smoking. Heatherton and Nichols asked their participants to write a narrative either

Nicotine, the most abused psychoactive substance, has been characterized as causing the most deadly epidemic of modern times (Stratton, Shetty, Wallace, & Bondurant, 2001). Besides causing serious health problems (e.g., cancer, heart disease, emphysema) costing billions of dollars annually (Andersen, Keller, & McGowan, 1999; Hurt et al., 1996; Klatsky, 1999; Rice, 1999), smoking is responsible for 25% to 33% of all deaths in the United States (Centers for Disease Control and Prevention, 1987/1997b; Hurt et al., 1996). Although the percentage of smokers in the U.S. population has declined from 42% in the mid-1960s (Stratton et al., 2001) to 25% in the mid-1990s (Centers for Disease Control and Prevention, 1997a; Wetter et al., 1998), the decline appears to be plateauing such that current smokers appeared to be more quit resistant (Stratton et al., 2001; Warner & Burns, 2003). Further, 2,000 youths start smoking cigarettes each day (Substance Abuse and Mental Health Administration, 2002). Thus, research on tobacco cessation is an important priority. Although several studies have shown that many smokers quit on their own (Carey, Snel, Carey, & Richards, 1989; Mariezcurrena, 1994; Marlatt, Curry, & Gordon, 1988), very little is known about the actual change process. An understanding of what factors contribute to and maintain the self-change process could be used to help smokers who have difficulty quitting. Researchers have suggested that because people “organize information about themselves, their social worlds, and their lives into narratives” (Baumeister, Stillwell, & Wotman, 1990, p. 994),

Todd M. Helvig, Linda Carter Sobell, Mark B. Sobell, and Edward R. Simco, Center for Psychological Studies, Nova Southeastern University. We gratefully acknowledge the editorial comments of Todd Heatherton. The research described in this article was conducted as part of Todd M. Helvig’s doctoral dissertation. Correspondence concerning this article should be addressed to Linda Carter Sobell, Center for Psychological Studies, Nova Southeastern University, 3301 College Avenue, Ft. Lauderdale, FL 33314. E-mail: [email protected] 219

BRIEF REPORTS

220

about an incident in which they had made a major and sudden change or about their inability (i.e., a failed attempt) to make a major and sudden change. Successful changers described themselves as having more control over their behavior than nonchangers did. Changers as compared with nonchangers also (a) described more consequential threats (e.g., health problems) and focal events (e.g., a conclusive decision) as reasons for change; (b) reported stronger negative affect (e.g., anxiety); (c) reported the importance and assistance of others (e.g., family members); and (d) reported significantly more statements relating to commitments to change (e.g., informing peers about the decision to change), more environmental change, and less ambivalence toward change. In the present study, smokers’ attempts to quit smoking cigarettes were the focal behavior in the change process. As in the Heatherton and Nichols (1994) study, participants’ narratives were subjected to content analysis to identify factors participants recalled as being central to the change process. On the basis of Heatherton and Nichols’s findings, it was hypothesized that narratives from smokers who had a maintained quit attempt compared with those who had quit but then returned to smoking would contain significantly more attributions of change to (a) self-control (e.g., “It is my responsibility to stop smoking”), (b) focal events (e.g., identification of events related to quitting), (c) affective events (e.g., “I worry about my health when I smoke”), and (d) social support (e.g., support and encouragement from friends and/or family to quit).

Method Participants Participants were recruited from visitors to a science center (Fort Lauderdale, FL) frequented mainly by tourists and local families. Using convenience samples is a common strategy for studying the change process with substance abusers (Klingemann et al., 2001; Sobell, Sobell, Toneatto, & Leo, 1993). Over several weeks, a sign near the science center entrance asked for volunteers who had quit smoking cigarettes or who were current smokers to participate in a short, confidential research study. No remuneration was offered, and the study was approved by the Nova Southeastern University Institutional Review Board. There were two different groups of participants: maintained quitters (i.e., ex-smokers; MQs; n ⫽ 59) or temporary quitters (i.e., current smokers; TQs; n ⫽ 47). Participants in the TQ group (a) had quit temporarily at least once in the year prior to the study, that is, quit for at least 1 day (Centers for Disease Control and Prevention, 1997a) but for no more than 90 days; (b) had smoked cigarettes for more than 1 year; and (c) were daily smokers (i.e., smoked more than 10 cigarettes per day). Participants in the MQ group (a) had smoked cigarettes for at least 1 year, (b) had been daily smokers before quitting (i.e., smoked more than 10 cigarettes per day), (c) had not been smoking cigarettes for at least 1 year, and (d) had a carbon monoxide (CO) reading of less than 8 ppm. An Ecolyzer (Vitalograph, Lenexa, KS) was used to verify that MQs were not current smokers (Hughes, Frederiksen, & Frazier, 1978; Vogt, Selvin, Widdowson, & Hulley, 1979).

Design and Procedure After signing an informed consent form, all participants completed a background and smoking history questionnaire (containing questions regarding, e.g., age, gender, number of years of smoking, number of cigarettes smoked per day). No information could be obtained about individuals who inquired about the study but declined to participate. All participants

were given one sheet of lined paper (8 1/2 ⫻ 11 in.) with the following instructions. For TQs, the instructions read, In as much detail as possible, describe your most serious attempt to quit smoking including what led up to your decision to quit. Note: most serious implies the time during which you most wanted to quit, and were successful for a period lasting one full day to no more than 3 full months. For MQs, the instructions read, “In as much detail as possible, describe the time you successfully quit smoking cigarettes including what led up to your decision to quit” (i.e., their most recent quit attempt). After completing their narratives, participants were debriefed about the study, and TQs were given an opportunity to request a referral for smoking treatment or to receive a free smoking cessation booklet.

Content Analysis of Narratives Differences in the content of the narratives written by MQs and TQs were evaluated using two types of content analysis: (a) dichotomous ratings of the presence or absence of different events and (b) computer content analysis of the frequency of events or words. The dichotomous rating procedures were similar to those used in previous studies (Baumeister et al., 1990; Heatherton & Nichols, 1994). Narratives were rated by two trained raters for the presence or absence of five change factors; four of those factors had been used in Heatherton and Nichols’s study (i.e., general self-control, focal events, strong negative affect, and general social support). A fifth factor (general emotional affect) was added to capture emotional events that were not limited to strong negative affect. Because dichotomous codings are more objective and yield higher interrater agreement compared with ratings on a continuum (Baumeister et al., 1990), they are the preferred way of evaluating micronarratives. Discrepancies between the two raters were reviewed by a third rater, who made the final coding determination. The interrater agreement rate was 83% (for TQ narratives, 82%; for MQ narratives, 85%), similar to rates reported in previous studies using micronarratives (Baumeister et al., 1990; Ellingstad et al., 2006; Heatherton & Nichols, 1994). For any given participant, any factor for which at least one coding was made was counted as the factor being present regardless of the total number of instances per factor. For each factor, the percentage of participants in each group whose narratives were judged to contain words representative of the factor was calculated. The second type of content analysis involved a code-and-retrieve software program (Miles & Huberman, 1994) called Textpack (Mohler & Zull, 1998) used in similar evaluations of substance abusers’ narratives (Klingemann, 1992; Sobell et al., 2001). Before the analysis, the link between text words, including synonyms, and concepts was defined in a content-analytic dictionary. Then computer coding and tagging automatically counted entries to the theoretical categories previously established (see Mochmann, 1999). The present categories and subcategories were based on a review of past qualitative and quantitative studies in this area (Klingemann et al., 2001) and a word dictionary (i.e., Natural Recovery Dictionary1) used in past self-change studies with substance abusers (Sobell et al., 2001). The words from the participants’ one-page narratives served as the raw text files. As in past studies, the raw text files were reduced by (a) dividing the narrative text into word segments or chunks (i.e., a single word or multiple words), (b) attaching codes to the chunks (e.g., “After my wife died of a smoking related cause, I started thinking about my own death if I continued to smoke” was coded as strong negative affect), and (c) displaying instances of coded chunks or combinations of chunks where patterns could be summarized into a meaningful understanding of the quit attempt. Similar analyses have been conducted with the narratives of

1

Copies of this dictionary are available from Linda Carter Sobell or Mark B. Sobell.

BRIEF REPORTS naturally recovered substance abusers in three different studies (Sobell et al., 2001).

221

Table 2 Percentage of Maintained Quitters’ and Temporary Quitters’ Narrative Stories Judged to Contain Different Change Factors

Results Although all participants’ narratives were included in the present analysis, for 6 participants (2 TQs and 4 MQs), an occasional word could not be deciphered and was omitted. Differences between MQs and TQs for smoking and demographic variables were analyzed using two-tailed t tests for continuous variables (e.g., age, number of years of smoking) and chi-square tests for nominal variables (e.g., gender, marital status). We used a multivariate analysis of variance (MANOVA) to analyze parametric variables (e.g., social support, cognitive components) within the narratives. The mean CO readings for both groups validated their selfreports of smoking (CO ⬎ 8 ppm) or not smoking (CO ⱕ 7 ppm): For MQs, M ⫽ 0.7 ppm, SD ⫽ 0.8 ppm; for TQs, M ⫽ 9.5 ppm, SD ⫽ 6.9 ppm; t(104) ⫽ 9.64, p ⬍ .001. Table 1 shows demographic and smoking variables for TQs and MQs and the results of significance tests of differences between the groups. The majority of participants were women, married, and white-collar employees and, on average, had some university education. Members of the MQ group were an average 10 years older than members of the TQ group and reported smoking an average of seven more cigarettes per day than those in the TQ group.

Dichotomous Codings of Micronarratives Differences in the presence of change factors in MQ and TQ narratives were evaluated using two-tailed Fisher’s exact tests, with the variables Change Factor (e.g., mention of self-control vs. no mention) ⫻ Smoking Status (MQ vs. TQ). Table 2 presents the percentages of MQ and TQ narrative stories judged to contain each of the change factors and significance test results. Because multiple tests were performed, a Bonferroni adjustment was applied to the alpha level, requiring a probability of less than .01 for statistical significance. The MQ group had a higher percentage of

% of narratives judged to contain different change factors

Change factors

Maintained quitters (n ⫽ 59)

Temporary quitters (n ⫽ 47)

p

General self-control Focal events General emotional affect Strong negative affect General social support

85 83 36 17 51

28 49 4 4 28

⬍.001 ⬍.001 ⬍.001 ⬎.05 .018

Note. Because the study was looking at the number of participants reporting different change factors, each change factor was coded as positive irrespective of the number of statements coded per change factor. The p values were for the Fisher’s exact test.

entries for all five change factors, and significant group differences were found for three of the change factors (general self-control, focal events, general emotional affect). The pattern of results for the changers (MQs) and nonchangers (TQs) was similar to that in the Heatherton and Nichols (1994) study (i.e., significantly more reasons were reported by changers as compared with nonchangers). Although two different change factors were used to evaluate affective components of the change process (general emotional affect, e.g., “I wanted to stop doing what I was doing”; strong negative affect, e.g., “I was very angry”), only the general emotional affect factor resulted in a significant difference ( p ⬍ .001).

Computerized Content Analysis of Narratives On the basis of the computerized Textpack analysis (Mohler & Zull, 1998), four major content areas (i.e., social support, cogni-

Table 1 Demographic and Smoking Variables Reported by Maintained Quitters and Temporary Quitters

Variable Demographic Age (in years) Education (in years) White collar Female Married Employed Smoking Years smoking cigarettes No. of serious quit attempts Cigarettes smoked per day Years quit Quit smoking ⱖ 1 day in the past year Smoking ⱕ 60 min on waking a

This was for the year prior to their recovery. ** p ⬍ .01.

Maintained quitters (n ⫽ 59)

Temporary quitters (n ⫽ 47)

M

SD

M

SD

49.9 14.5

17.3 3.2

39.3 13.9

15.6 2.6

%

62.7 74.6 62.7 44.1 20.4 4.0 21.8a 11.9

14.3 5.1 13.0 11.6

t(104)

␹2(1)

⫺3.31** ⫺1.12 55.3 59.6 57.4 53.2

18.9 3.6 15.2 100.0a 68.3a

%

0.59 2.70 0.30 0.87 ⫺0.51 ⫺0.43 ⫺3.03**

16.1 4.9 9.3 100.0 64.7

0.00 0.18

BRIEF REPORTS

222

tive, affective, and physical health) emerged with relatively high count frequencies for both groups. With one exception (i.e., physical health), large word chunks emerged that allowed the content areas to be divided into anti–smoking cessation and pro–smoking cessation content. To help readers understand the content areas, examples for each change factor are listed in Table 3, which shows the adjusted means and statistical tests of the Textpack analysis of counts of words and phrases from the narratives of the MQ and TQ groups. Because the differences of interest between the MQ and TQ groups focused on specific change factors and not on a linear combination of the factors, multiple univariate analyses were used to evaluate differences (Huberty & Morris, 1989). Because the groups were known to differ in age and age differences could be related to certain change factors such as physical health factors, analyses of covariance (ANCOVAs) were used with age as the covariate. Change factor scores were the number of coded entries in each category. Results from the ANCOVAs for the main change factors of social support, cognitive, affective, and physical health factors are shown in Table 3 and reveal the groups differed significantly only for the affective factor. Adjusted means for the groups, also shown in Table 3, show that the MQ group had more comments coded for affect than did the TQ group. For three of the four change factors (social support, cognitive, and affective), it was possible to categorize the entries for anti– smoking cessation (negative) and pro–smoking cessation (positive) content. When further coded in this manner, several further significant differences were found. As shown in Table 3, a pattern was noted in that for each factor, the MQ group had higher adjusted means than did the TQ group, indicating that the MQ group’s narratives contained significantly more pro–smoking cessation statements than the TQ group’s narratives did. For the social support and affective factors, it was found that the members of the TQ group made more anti–smoking cessation comments than members of the MQ group did.

Discussion In this exploratory study, we used attempts to quit smoking cigarettes as a focal behavior to understand factors related to the change process for smokers who had quit smoking for a year or more (i.e., MQs) and compared their experiences with those of smokers who had made a serious quit attempt (i.e., quit for 1 to 90 days) but had resumed smoking prior to being interviewed (i.e., TQs). In autobiographical narratives, participants commented on what led up to their last quit attempt (MQs) or most serious quit attempt (TQs). The study’s methodology, which was similar to that used by Heatherton and Nichols (1994), extended the evaluation of the change process to a specific behavior, smoking cessation. The analysis of participants’ narratives suggested that several change factors seemed to facilitate (i.e., for MQs) or impede (i.e., for TQs) the change process. Like Heatherton and Nichols (1994), we found that significantly more maintained changers (i.e., MQs) wrote statements in their narratives about general self-control, focal events, and general social support compared with nonchangers (i.e., TQs). However, the computerized content analysis using age as a covariate found only one significant difference between groups for the general factors—MQs made significantly more affective comments than did TQs. This suggests that some of the differences between groups as judged by raters might reflect age rather than quitting status differences. The computerized content analysis yielded several other clinical implications when the valence (pro–smoking cessation, anti– smoking cessation) of narrative statements was coded in addition to the content. For the three categories that could be valence coded (social support statements, cognitive statements, affective statements), in all but one case, there were significant differences involving the MQ group making more pro–smoking cessation statements and fewer anti–smoking cessation statements than the TQ group. The one exception was for cognitive anti–smoking cessation statements, although the direction of group differences was consistent with the other variables. These findings are consis-

Table 3 Computerized Content Analysis Adjusted Means of Word Counts from the Maintained and Temporary Quitters’ Narratives and Text Examples of Change Factors

Change factors Social support Anti–smoking cessation (“I always smoke with my friends at the bar.” “My daughters smoke, so it’s hard not to when I’m with them.”) Pro–smoking cessation (“My friends asked me to quit smoking.” “My family urged me to stop.”) Cognitive Anti–smoking cessation (“Temptation set in, and the next thing I knew I was smoking.” “I didn’t want to stop smoking.”) Pro–smoking cessation (“I just decided not to smoke.” “I know what smoking will do to me.”) Affective Anti–smoking cessation (“I really missed smoking.” “I enjoyed smoking.”) Pro–smoking cessation (“I was disgusted [by smoking].” “I felt so powerless.”) Physical health (“It became hard for me to inhale.” “I got emphysema.”) * p ⬍ .05.

** p ⬍ .01.

Maintained quitters Adj. M (n ⫽ 59)

Temporary quitters Adj. M (n ⫽ 47)

F(1, 103)

1.06

0.90

0.56

0.07

0.49

10.67**

0.97 0.93

0.42 0.70

10.70** 1.24

0.11

0.27

2.90

0.81 0.52 0.01 0.50 0.99

0.45 0.21 0.14 0.08 0.71

4.89* 4.14* 4.97** 8.42** 1.76

BRIEF REPORTS

tent with those of other studies and not surprising. For example, Mermelstein, Cohen, Lichtenstein, Baer, and Kamarck (1986) found that anti–smoking support from significant others was important in helping smokers refrain from smoking. Likewise, cigarette smokers show worse outcomes when friends accept their smoking (reviewed in Havassy, Hall, & Wasserman, 1991). However, the relationship of such comments to smoking status only became apparent when statements were subcategorized by smoking cessation valence. Examining smokers’ reports of their quit attempts with respect to anti–smoking cessation and pro–smoking cessation content, therefore, is important, as different findings emerged when valence was considered compared with when the narratives were evaluated without regard to valence. This study has several limitations. First, because a convenience sample was used (i.e., volunteers at a local science center), generalization of findings to all cigarette smokers awaits further study. In addition, about two thirds of the participants were women, and the average education level was more than 2 years beyond high school. Second, although participants wrote their narratives in a public setting and used a single sheet of paper, there is no reason to think these conditions differentially impacted the two groups. Third, the maintained quitters were significantly older than the temporary quitters were. It is possible that older smokers develop more health problems and that this accounts for more reports of physical health factors and an increased likelihood of cessation. Fourth, as in many smoking cessation studies, although verification of the length of smoking cessation (i.e., number of years) was not possible, some evidence suggests that participants’ reports have face value. The CO readings for both groups corroborated their current reported smoking status, and the findings from their narratives are consistent with past studies of change factors. Fifth, because this study measured participants’ recall of factors related to their attempts at smoking cessation, their reports are subject to possible memory and attribution problems. One way to address this would be to prospectively gather information from smokers as they attempt to quit and at different times in the cessation process. Such research, however, is costly and takes several years to conduct. In the interim, retrospective studies such as the current investigation can be used to suggest possible research directions. Last, regardless of whether the participants’ recollections are technically accurate, their perceptions of their reasons for change are likely to impact the maintenance of change and be related to relapse risk. The present findings have implications for new intervention and self-change strategies and for developing a better understanding of factors that initiate and maintain smoking cessation (e.g., increasing positive social support in the smoker’s environment, identifying and developing ways to avoid cues that contribute to further smoking). In this regard, although personal accounts of behavior change might be subject to attributional bias, a caution put forth by Heatherton and Nichols (1994) is relevant: Reasons provided by changers, retrospective or not, are real, and such beliefs are likely to “have a powerful influence on future attitudes and behaviors, especially attitudes and behaviors aimed at trying to achieve or maintain life change” (p. 673). Studying the change process from the changers’ subjective viewpoint adds a dimension to studies of the recovery process that is neglected in most treatment research and suggests interesting variables for future study. For example, a smoking cessation program could be presented to participants in

223

which factors that successful quitters reported as being important to their recovery are emphasized.

References Andersen, S., Keller, C., & McGowan, N. (1999). Smoking cessation: The state of the science. The utility of the transtheoretical model in guiding interventions in smoking cessation. Online Journal of Knowledge Synthesis for Nursing, 6, Article 2. Baumeister, R. F. (1996). The crystallization of discontent in the process of major life change. In T. F. Heatherton & J. L. Weinberger (Eds.), Can personality change? (pp. 281–297). Washington, DC: American Psychological Association. Baumeister, R. F., & Stillwell, A. M. (1992). Autobiographical accounts, situational roles, and motivated biases: When stories don’t match. In A. L. Weber, T. L. Orbuch, & J. H. Harvey (Eds.), Attributions, accounts, and close relationships (pp. 52–70). New York: SpringerVerlag. Baumeister, R. F., Stillwell, A., & Wotman, S. R. (1990). Victim and perpetrator accounts of interpersonal conflict: Autobiographical narratives about anger. Journal of Personality and Social Psychology, 59, 994 –1005. Bottorff, J. L., Johnson, J. L., Irwin, L. G., & Ratner, P. A. (2000). Narratives of smoking relapse: The stories of postpartum women. Research in Nursing & Health, 23, 126 –134. Burman, S. (1997). The challenge of sobriety: Natural recovery without treatment and self-help groups. Journal of Substance Abuse, 9, 41– 61. Carey, M. P., Snel, D. L., Carey, K. B., & Richards, C. S. (1989). Self-initiated smoking cessation: A review of the empirical literature from a stress and coping perspective. Cognitive Therapy and Research, 13, 323–341. Centers for Disease Control and Prevention. (1997a, December 26). Cigarette smoking among adults—United States, 1995. Morbidity and Mortality Weekly Report, 46, 1217–1220. Centers for Disease Control and Prevention. (1997b, May 23). Perspective in disease prevention and health promotion smoking-attributable mortality and years of potential life lost—United States, 1984. Mortality and Morbidity Weekly Report, 46, 444 – 451. (Original work published October 30, 1987) Ellingstad, T. P., Sobell, L. C., Sobell, M. B., Eickleberry, L., & Golden, C. J. (2006). Natural recovery from cannabis abuse: Factors associated with self-change. Addictive Behaviors, 31, 519 –530. Gergen, K. J., & Gergen, M. M. (1998). Narrative and the self as relationship. In L. Berkowtiz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 17–56). San Diego, CA: Academic Press. Hanninen, V., & Koski-Ja¨nnes, A. (1999). Narratives of recovery from addictive behaviours. Addiction, 94, 1837–1848. Harvey, J. H., Weber, A. L., & Orbuch, T. L. (1990). Interpersonal accounts: A social psychological perspective. Cambridge, MA: Basil Blackwell. Havassy, B. E., Hall, S. M., & Wasserman, D. A. (1991). Social support and relapse: Commonalities among alcoholics, opiate users, and cigarette smokers. Addictive Behaviors, 16, 235–246. Heatherton, T. F., & Nichols, P. A. (1994). Personal accounts of successful versus failed attempts at life change. Personality and Social Psychology Bulletin, 20, 664 – 675. Huberty, C. J., & Morris, J. D. (1989). Multivariate analysis versus multiple univariate analyses. Psychological Bulletin, 105, 302–308. Hughes, J. R., Frederiksen, L. W., & Frazier, M. (1978). A carbon monoxide analyzer for measurement of smoking behavior. Behavior Therapy, 9, 293–296. Humphreys, K. (2000). Community narratives and personal stories in Alcoholics Anonymous. Journal of Community Psychology, 28, 495– 506.

224

BRIEF REPORTS

Hurt, R. D., Offord, K. P., Croghan, I. T., Gomezdahl, L., Kottke, T. E., Morse, R. M., et al. (1996). Mortality following inpatient addictions treatment: Role of tobacco use in a community-based cohort. Journal of the American Medical Association, 275, 1097–1103. Kaufman, S. A. (1986). The ageless self: Sources of meaning in late life. New York: Meridian. Klatsky, A. L. (1999). Moderate drinking and reduced risk of heart disease. Alcohol Health & Research World, 23, 15–23. Klingemann, H. K. H. (1991). The motivation for change from problem alcohol and heroin use. British Journal of Addiction, 86, 727–744. Klingemann, H. K. H. (1992). Everyday definitions of deviant behavior: Computer-assisted content analysis of lay concepts of alcohol and drug problems, delinquency and youth problems. ZUMA–Publikationen, 24, 105–129. Klingemann, H., Sobell, L., Barker, J., Blomqvist, J., Cloud, W., Ellingstad, T. P., et al. (2001). Promoting self-change from problem substance use: Implications for policy, prevention and treatment. Boston: Kluwer Academic. Mariezcurrena, R. (1994). Recovery from addictions without treatment: Literature review. Scandinavian Journal of Behaviour Therapy, 23, 131–154. Marlatt, G. A., Curry, S., & Gordon, J. R. (1988). A longitudinal analysis of unaided smoking cessation. Journal of Consulting and Clinical Psychology, 56, 715–720. McAdams, D. P. (1985). Power, intimacy, and the life story: Personological inquires into identity. Homewood, IL: Dorsey. McAdams, D. P. (1994). Can personality change? Levels of stability and growth in personality across the life span. In T. Heatherton & J. Weinberger (Eds.), Can personality change? (pp. 299 –313). Washington, DC: American Psychological Association. McIntosh, J., & McKeganey, N. (2000). Addicts’ narratives of recovery from drug use: Constructing a non-addict identity. Social Science & Medicine, 50, 1501–1510. Mermelstein, R., Cohen, S., Lichtenstein, E., Baer, J. S., & Kamarck, T. (1986). Social support and smoking cessation and maintenance. Journal of Consulting and Clinical Psychology, 54, 447– 453. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage. Mochmann, E. (1999, August). Social science data analysis and collection. Paper presented at the 32nd Essex Summer School, Essex, England. Mohler, P. P., & Zull, C. (1998). Textpack (Version for Windows 95/NT)

[Computer software]. Mannheim, Germany: Zentrum fu¨r Umfragen, Methoden, Analysen. Parry, O., Fowkes, F. G. R., & Thomson, C. (2001). Accounts of quitting among older ex-smokers with smoking-related disease. Journal of Health Psychology, 6, 481– 493. Rice, D. P. (1999). Economic cost of substance abuse, 1995. Proceedings of the Association of American Physicians, 111, 119 –125. Rosenberg, S. (1988). Self and others: Studies in social personality and autobiography. In L. Berkowtiz (Ed.), Advances in experimental social psychology (pp. 57–95). San Diego, CA: Academic Press. Sobell, L. C., Klingemann, H., Toneatto, T., Sobell, M. B., Agrawal, S., & Leo, G. I. (2001). Alcohol and drug abusers’ perceived reasons for self-change in Canada and Switzerland: Computer-assisted content analysis. Substance Use and Misuse, 36, 1467–1500. Sobell, L. C., Sobell, M. B., Toneatto, T., & Leo, G. I. (1993). What triggers the resolution of alcohol problems without treatment? Alcoholism: Clinical and Experimental Research, 17, 217–224. Stratton, K., Shetty, P., Wallace, R., & Bondurant, S. (2001). Clearing the smoke: The science base for tobacco harm reduction— executive summary. Tobacco Control, 10, 189 –195. (Reprinted from Clearing the smoke: The science base for tobacco harm reduction, pp. 1–18, by K. Stratton, P. Shetty, R. Wallace, & S. Bondurant [Eds.], 2001, Washington, DC: National Academies Press) Substance Abuse and Mental Health Administration. (2002). National Household Survey on Drug Use and Health: Main findings 2002. Rockville, MD: U.S. Department of Health and Human Services. Vogt, T. M., Selvin, S., Widdowson, G., & Hulley, S. B. (1979). Expired air carbon monoxide and serum thiocyanate as objective measures of cigarette exposure. American Journal of Public Health, 67, 545–549. Warner, K. E., & Burns, D. M. (2003). Hardening and the hard-core smoker: Concepts, evidence, and implications. Nicotine & Tobacco Research, 5, 37– 48. Wetter, D. W., Fiore, M. C., Gritz, E. R., Lando, H. A., Stitzer, M. L., Hasselblad, V., et al. (1998). The Agency for Health Care Policy and Research Smoking Cessation Clinical Practice Guideline: Findings and implications for psychologists. American Psychologist, 53, 657– 669.

Received May 28, 2004 Revision received May 20, 2005 Accepted May 25, 2005 䡲