PII S0306-4603(97)00072-5 BRIEF REPORT

Pergamon Addictive Behaviors, Vol. 23, No. 4, pp. 537 541, 1998 Copyright 1998 Elsevier Science Ltd Printed in the USA. All rights reserved 0306-4603/98 $19.00.00 PII S0306-4603(97)00072-5 BRIEF REPORT NATURAL RESOLUTION OF ALCOHOL PROBLEMS WITHOUT TREATMENT: ENVIRONMENTAL CONTEXTS SURROUNDING THE INITIATION AND MAINTENANCE OF STABLE ABSTINENCE OR MODERATION DRINKING MICHELE PUKISH KING and JALIE A. TUCKER Auburn University Abstract Environmental contexts surrounding the natural resolution of alcohol problems were investigated using untreated former problem drinkers who had maintained stable abstinence (n 18) or moderation drinking (n 17) for more than 2 years (M 7.1 years resolved). Untreated active problem drinkers (n 17) served as controls. During structured interviews, events were assessed retrospectively over a 4-year period that spanned the 2 years before and the 2 years after stable behavior change was initiated or over a matched recall interval for controls. Collaterals verified participant reports. Compared to nonresolved controls, both resolved groups reported improved life circumstances during the first year of maintenance, which concurs with treatment outcome studies showing that the circumstances during the posttreatment interval influence long-term outcomes. These positive changes diminished over time, however, among moderation drinkers only as their drinking remained normalized, which suggests that tolerating life changes during maintenance is a feature of successful moderation. 1998 Elsevier Science Ltd Emphasis on treatment outcome research in the alcohol field has deterred investigation of natural influences on the drinking behavior change process (e.g., Sobell, Sobell, & Toneatto, 1992; Tucker, Vuchinich, & Gladsjo, 1994). Such studies are important because most problem drinkers never enter treatment, many who do receive treatment do not sustain positive behavior change, and moderation drinking outcomes are more common among natural than treatment-assisted recoveries (e.g., Sobell, Cunningham, & Sobell, 1996). Research on natural resolutions may suggest improved interventions to facilitate the natural contingencies that promote successful behavior change and may help identify variables to guide the selection of drinking goals. Environmental contextual variables have been shown to influence drinking patterns after alcohol treatment (reviewed by Tucker, Vuchinich, & Gladsjo, 1990 1991), with positive behavior changes being associated with improved life circumstances and resources during the posttreatment interval. Similar improvements after initial behavior change have been associated with natural resolutions (e.g., Tucker et al., 1994). This suggests that a common molar, or temporarily extended, environmental context surrounds resolutions achieved through different pathways (Tucker, Vuchinich, & Puk- This research was based on a dissertation completed by Michele King under the direction of Jalie Tucker. It was supported in part by a grant-in-aid award to Jalie Tucker from the Office of the Vice President for Research at Auburn University. Manuscript preparation was supported by Grants R01 AA08972 and K02 AA00209 to Jalie Tucker from the National Institute on Alcohol Abuse and Alcoholism. Portions of the research were presented at the 102nd annual convention of the American Psychological Association, Los Angeles, August 1994. Requests for reprints should be sent to Jalie Tucker, Department of Psychology, 226 Thach Hall, Auburn University, AL 36849; E-mail: tuckeja@mail.auburn.edu 537

538 M. P. KING and J. A. TUCKER ish, 1995) and that improved circumstances after initial change may support and reinforce continued maintenance efforts. Much of this research focused on abstinent resolutions, and the generality of findings to stable moderation outcomes is not well established. Therefore, the present study investigated environmental events during a multiyear period surrounding the attainment of stable abstinence or moderation drinking by untreated problem drinkers. Studying the contexts that surround successful resolutions, especially those that exist during maintenance, also seems warranted because few reliable predictors of abstinent versus moderation outcomes have been identified from drinking history, demographic, and psychopathology variables (Rosenberg, 1993). METHOD The method was identical to Tucker et al. (1994, 1995), who investigated abstinent resolutions achieved with and without interventions. Pukish (1994) and King and Tucker (1997) described the present sample and procedures in detail. Untreated problem drinker participants with a minimum 5-year problem history (M 14.02 years, SD 6.89) were recruited from the community using media solicitations. They varied in their current drinking status: (a) continuously abstinent for more than 2 years (n 18); (b) moderation drinking without problems for more than 2 years, with or without intermittent periods of abstinence (n 17); or (c) active problem drinking (n 18). The nonresolved (NR) control group served to ascertain which patterns of events were uniquely associated with resolution and were not simply common occurrences in the lives of problem drinkers (Sobell et al., 1992). The resolved abstinent (RA) and resolved nonabstinent (RNA) participants had similar, lengthy resolutions (M 7.09 years, SD 3.64). During structured interviews, event occurrences were assessed retrospectively over a 4-year period that began 2 years before resolved participants initiated stable abstinence or moderation and then continued for 2 years of maintenance. NR participants recalled events over a similar 4-year retrospective period during which time they were drinking abusively. Events were assessed using the Life Events Questionnaire (LEQ; see Tucker et al., 1995), which assesses events in nine functional categories (work, residence, intimate relations, family relations, friendship/social, finances, legal, health problems, health habits). For each event reported, participants indicated the date(s) of occurrence and whether it had a positive or negative effect. Phone interviews were conducted with collaterals for 73.1% of participants. As described in King and Tucker (1997), the collateral reports supported the integrity of participants reports. RESULTS Events were summed for each year for total positive and total negative events and for each functional category on the LEQ collapsing over positive and negative events. The low frequency of events in some years precluded separate analyses of positive and negative events within each category. The annual event frequencies were analyzed using 3 (Group) 4 (Year) repeated measures analyses of variance (ANOVAs). Of primary interest were Group Year interactions that would reveal improved circumstances during resolution maintenance. Significant interactions were examined using simple main effects tests, with alpha distributed among the comparisons according to

Contexts surrounding natural resolution 539 whether the effects were predicted or were being evaluated post hoc. Note that the RA and RNA participants initiated stable resolutions at the end of Year 2. The ANOVA for total positive events yielded only a conceptually uninteresting main effect for years, F(3, 147) 3.74, p.02, indicating that participants in all groups tended to report more positive events over time. Of greater interest was a near significant Group Year interaction for total negative events, F(6, 147) 2.14, p.052, shown in Figure 1, that supported the hypothesized pattern of events for RA and NR participants. Comparisons showed that negative events decreased from the prethrough the postresolution interval for RA participants (p.05) and increased over time for NR participants (p.01) and that the two groups differed significantly during Years 3 and 4 (ps.01), when RA participants reported fewer negative events. Fig. 1. Group Year interaction for total negative events as assessed by the Life Events Questionnaire. The figure shows the mean event occurrences per year for each drinking status group. Resolved abstinent and resolved nonabstinent participants had initiated stable resolutions at the end of Year 2.

540 M. P. KING and J. A. TUCKER Although the RNA participants did not evidence significant change over years, their annual mean frequencies were similar to those of RA participants, except during Year 4 when they reported increased negative events that approached the mean for NR participants; during Year 3, their mean frequency was significantly lower than the mean for NR participants (p.01). The ANOVAs for the LEQ functional categories revealed no significant interaction effects in the legal, work, finances, residential, health habit, and social categories. Significant Group Year interactions were obtained for health problems, F(6, 147) 2.26, p.041; family relations, F(6, 147) 2.57, p.022; and intimate relations, F(6, 147) 2.29, p.039. The interaction for health events was similar to the interaction for total negative events. Both resolved groups reported fewer health problems during Year 3 compared to NR participants (ps.05), who reported more events across all years. RA participants continued to report reduced events during Year 4, whereas RNA participants reported increased health events that approached the mean for NR participants. For events in the intimate relations and family relations categories, both the RA and NR groups showed significant change over time (ps.01), whereas the RNA group did not. Similar to their pattern for total negative events, events in both categories increased over time for NR participants. For RA participants, events in both categories decreased during Year 4 compared to previous years; in addition, from Years 2 to 3, there was a shift from predominately negative to positive events in both categories, even though the mean levels were similar in these years. Although the RNA participants did not show significant change over time in either category, their pattern of intimate relations events was similar to that observed for health problems (i.e., decreased events from Year 2 to Year 3 followed by an increase in Year 4). For family relations events, their annual mean frequencies were low and similar across all years. Finally, following Tucker et al. (1995), raters categorized participants reports of their reasons for resolution and influences on maintenance ( 90% agreement). Similar to past research (e.g., Tucker et al., 1994, 1995), all reasons for resolution were negative and typically developed over a lengthy period of time before active behavior change (68.6% of participants). Health problems were cited most frequently (37.1%), and smaller percentages of participants cited reasons in all other LEQ categories except for social relations and residential changes. With respect to resolution maintenance, family members (37.1%) and religious involvement (31.4%) were cited most frequently. The only group difference was that RA participants endorsed changes in will power (22.2%) to a greater degree than did RNA participants (0%), 2 (1) 4.26, p.05. DISCUSSION Consistent with research on treatment-assisted and natural behavior change in problem drinkers (e.g., Tucker et al., 1995), the event findings suggest that improved life circumstances after initial behavior change are associated with long-term success and may function to support and reinforce continued maintenance efforts. Compared to the NR participants who reported increased negative events over time, both resolved groups reported decreased health problems and decreased total negative events during the first year of maintenance. This pattern generally replicates earlier research on natural resolutions that involved abstinence (e.g., Tucker et al., 1994) and extends it to those that involved moderation drinking, but only through the first year of maintenance.

Contexts surrounding natural resolution 541 An unexpected but intriguing finding was that the initially improved life circumstances experienced by both resolved groups diminished over time for moderation drinkers but not for abstinent participants, whose lives continued to stabilize during the second year of maintenance. A conventional interpretation would be that any drinking by former problem drinkers produced life-health stressors and problems that would increase the risk of relapse over time. However, this had not happened because the RNA participants had maintained moderation for many years. An alternative view is that, as their drinking remained normalized and less central as a life problem, the moderation drinkers were increasingly able to tolerate some instability and change without resuming problem drinking; i.e., they were resilient and adaptable in the face of typical variations in living conditions. By comparison, the environments of abstainers were increasingly uneventful, whereas the greatest instability in life circumstances occurred among the active problem drinkers. The causal status and direction of these environmental differences in relation to different drinking patterns and outcomes obviously cannot be determined from this retrospective, correlational study, and the modest sample consisting mainly of white men further qualifies the findings. Nevertheless, the data raise interesting questions for future research about the contexts that support different drinking goal choices and patterns of behavior change and how individual differences in tolerating life changes may affect long-term drinking outcomes. The findings are less informative about preresolution circumstances that motivate initial behavior change or influence drinking goal choices because the environments of the resolved groups did not differ much until the second year of maintenance. Given the ubiquity of resolution attempts that end in relapse, perhaps it is not surprising to find that the variables influencing long-term success appear to reside primarily in what happens after initial behavior change. Successful change is thus probably best viewed as a lengthy process that extends well beyond the initial act of quitting problem drinking and one that is dependent on the surrounding environmental contexts. Research on natural resolutions offers an opportunity to study the contexts and processes involved in successful change independent of any influence of treatment. REFERENCES King, M. P., & Tucker, J. A. (1997). Achieving stable moderation drinking or abstinence during the natural resolution of alcohol problems without treatment: Behavior change patterns and strategies. Manuscript submitted for review. Pukish, M. M. (1994). Environmental contexts surrounding natural recovery from alcohol problems without treatment: A comparison of abstinent and nonabstinent recoveries (Doctoral dissertation, Auburn University, 1994). Dissertation Abstracts International, 55, 3000B. Rosenberg, H. (1993). Prediction of controlled drinking by alcoholics and problem drinkers. Psychological Bulletin, 113, 129 139. Sobell, L. C., Cunningham, J. A., & Sobell, M. B. (1996). Recovery from alcohol problems with and without treatment: Prevalence in two population studies. American Journal of Public Health, 86, 966 972. Sobell, L. C., Sobell, M. B., & Toneatto, T. (1992). Recovery from alcohol problems with and without treatment. In N. Heather, W. R. Miller, & J. Greeley (Eds.), Self-control and addictive behaviors (pp. 198 242). New York: Maxwell MacMillan. Tucker, J. A., Vuchinich, R. E., & Gladsjo, J. A. (1990 1991). Environmental influences on relapse in substance use disorders. International Journal of the Addictions, 25, 1017 1050. Tucker, J. A., Vuchinich, R. E., & Gladsjo, J. A. (1994). Environmental events surrounding natural recovery from alcohol-related problems. Journal of Studies on Alcohol, 55, 401 411. Tucker, J. A., Vuchinich, R. E., & Pukish, M. M. (1995). Molar environmental contexts surrounding recovery from alcohol problems by treated and untreated problem drinkers. Experimental and Clinical Psychopharmacology, 3, 195 204.